Summary

In the public internet space, numerical patterns periodically emerge that quickly acquire the status of "hidden knowledge" because they connect highly charged historical dates through a simple, easily reproducible operation. One such pattern is the so-called "68 pattern," in which the start dates of World War I, World War II and the full-scale Russian invasion of Ukraine yield the same value when summing the day, month and the two two-digit parts of the year. At first glance, this looks like internet numerology. At second glance, the question arises whether something more interesting is at play: not proof of a mystical regularity, but at least a testable statistical and historical phenomenon.

The present text examines the problem not as grounds for superstition, but as a methodological test. The first task is to determine exactly how the score is calculated, what counts as the start of a war, which cases are admissible and how calendars are normalized when comparing dates across different centuries. The second task is to see what happens when the pattern is applied not only to the three popular cases, but to a working corpus of wars with reliably established exact start dates in day-month-year format. The third task is to understand whether the four cases with score 68 can be described merely as a curiosity, or whether they form a denser historical cluster with particularly high systemic significance.

The result is twofold. On one hand, the pattern is not sufficient to claim the existence of a proven law, a discovered "code" of history or an external marking upon wars. On the other hand, in the precisely dated and verifiable corpus used here, score 68 appears not once but four times, and specifically upon four wars or war onsets with extraordinarily high historical density: the Italo-Turkish War, World War I, World War II and the full-scale Russian invasion of Ukraine. This is not proof of metaphysics, but it is unusual enough to deserve serious, disciplined and unhurried examination.

Main Text

1. Why it is worth writing such an article at all

Internet hype usually operates through two simultaneous actions: it dramatizes the connection and hides the methodology. In this specific case, the 68 pattern spreads as an almost self-proving suggestion: three dates are seen, a threefold addition is performed, the same number is obtained, and the mind immediately enters a mode of symbolic alarm. If the subject were random, such a pattern would remain an interesting detail. But when we are talking about wars with enormous human, civilizational and moral consequences, even an apparently absurd numerical hypothesis deserves verification precisely in order to separate the observable from the imagined.

From the perspective of "History of the Future," the problem is deeper than numerology itself. Modern civilization does not suffer only from a shortage of information, but from a shortage of a disciplined framework in which signal and noise can be separated without hastily dismissing the strange and without hastily sacralizing the coincidence. If every unusual correlation is automatically declared nonsense, the system becomes deaf to early indicators. If every correlation is automatically declared proof, the system dissolves into conspiratorial thinking. Therefore, the true subject of this text is not just the number 68, but the question of how a mature analytical culture should process unusual historical coincidences.

There is yet another reason why the topic is important. Many of the most dangerous errors in the age of informational hyperspeed arise not when people lack data, but when they have a small, visually powerful pattern and accept it as an explanation. This mechanism is characteristic of both conspiratorial and technocratic cultures. Both love the short code. Both want events to fold into a quick formula. This text refuses both comforts. It takes the pattern seriously, but not as an answer - rather as an occasion to build a working methodology for verification.

Therefore, the value of this text lies not in accepting the pattern as truth, but in testing whether beneath an apparently noisy signal there is an observable historical structure.

2. What exactly is the "68 pattern" and how the score is calculated

In mathematical terms, there is nothing mystical here. This is simply an arbitrarily defined function applied to a calendar date. However, the arbitrariness does not automatically invalidate the result. In science, many useful indices also begin as conditional constructions. The question is not whether the function is "natural," but whether it is clearly defined, fully reproducible and applied consistently. If we use the same rule every time, if we do not change the year format depending on the case and if we do not substitute the start date to achieve a desired result, then at least we have a test, not a trick.

Another important distinction must be made here. Score is not a "cause," not a "meaning," not an "energy" of the event and not proof of a hidden code of history. Score is a numerical projection of a date. Anything beyond that must be additionally proven. If four wars yield 68, this means only that four start dates, processed with the same function, coincide in result. Any claim beyond that must pass through a historical and probabilistic sieve, otherwise it becomes metaphysics disguised as analysis.

3. Why the phrase "all wars of humanity" cannot be used carelessly

The biggest mistake in this type of investigation would be to say: "we applied the pattern to all wars in history." Such a sentence sounds powerful, but it is scientifically indefensible if it is not clear what exactly "all" means, what a "war" is and what a "known start date" is. A large portion of ancient and medieval conflicts are dated only by year, season, reign, campaign or approximate interval. For some of them, even the question "what is the first day of the war" is unsolvable without serious conditionality. This means that a pattern that requires a day and a month cannot honestly be applied to the entire military archive of humanity.

Precisely for this reason, the present text does not work with a claim to absolute exhaustiveness, but with corpus discipline, and therefore a stricter operational definition is introduced here. Admissible for calculation is only that war or military phase for which a reliable start date in day-month-year format can be established from a stable historical source, and in such a way that the beginning is not purely arbitrarily chosen from the multitude of possible episodes. If a given conflict has only a yearly or monthly dating, it is not included in the numerical calculation. If there are two legitimate beginnings - for example, a broader war and a later full-scale phase - this is explicitly noted and only the phase that is clearly named is calculated.

Therefore, the present article does not claim metaphysical completeness, but methodological correctness. It works not with the fantasy of the absolute archive, but with a precisely dated working corpus. This is less impressive, but much more reliable. From the perspective of Gen FUTURE, this is a key principle: better a limited truth with clear boundaries than a global claim that collapses at the first check.

4. The calendar is not a minor detail but a condition for the meaningfulness of the pattern

When comparing dates across long historical periods, the calendar ceases to be a technical note and becomes a fundamental problem. The Julian and Gregorian calendars do not coincide, and in different countries the transition occurred at different moments. If one calculates a pattern on old dates without specifying whether the Old Style, New Style or a unified projection is used, one is actually not working with the same time system. This means that the score can change not because the event is different, but because the calendar framework is different.

For this reason, the principle of normalization to the proleptic Gregorian calendar is adopted here - that is, the Gregorian calendar projected backward in time for the purposes of systematic comparison. This does not mean that ancient and early modern people lived in this calendar. It only means that when we want comparable machine processing of dates, we need a unified coordinate system. Without it, we get a mix of dates that look uniform but are not.

This step is particularly important because it does not allow the pattern to "win" from calendar fog. If someone claims that a given early war also yields 68 but uses the Old Style for one date and the New Style for another simultaneously, the result is analytically invalid. Therefore, in this text ancient and early cases are not forced but constrained. Where the calendar conversion and the dating itself are not sufficiently stable, the score is not calculated. This is not a weakness of the method. It is its minimal scientific integrity.

5. The beginning of a war is not just a date but a historical decision

Another critical difficulty is that "beginning" can mean different things. There is a declaration of war, a first battle, first casualties, first border crossing, first mass mobilization, first phase of open invasion, and in some conflicts there is a long transition between a hybrid phase and open war. If this problem is not addressed, the pattern becomes a function of editorial taste. Some will choose the diplomatic date, others the operational one, still others the symbolic one.

Here a hierarchy of choice is adopted. When there is a widely accepted and clearly stated start date of the war in a primary historical source, it is used. When there are different levels of the conflict, as with the Russo-Ukrainian war, they are explicitly distinguished: the broader war begins in February 2014, while the full-scale invasion begins on 24.02.2022. If we talk about the score of "the war in Ukraine" without specification, we risk a methodological substitution. If we talk about the score of "the full-scale Russian invasion of Ukraine," we already have a clearly formulated subject.

This distinction also has moral significance. A numerical pattern must not be allowed to blur the question of who started what and in what mode. Italy is the aggressor in the Italo-Turkish War. Nazi Germany is the aggressor whose attack on Poland starts World War II in Europe. Russia is the aggressor in the full-scale invasion of Ukraine in 2022. In the case of World War I, the picture is more complex due to the system of alliances and escalations, but even there the specific starting act is clear: Austria-Hungary declares war on Serbia on 28.07.1914. Score does not neutralize responsibility. It operates on a date, not on the moral evaluation of the action.

6. How the methodology for searching and sorting the pattern is applied

Once the formula and the rules of admissibility are fixed, the calculation itself is simple. For each admissible war, the start date is extracted, the date is recorded in a unified format, the year is split into two two-digit parts and the score is calculated. Then the cases are sorted by score in ascending order, and in case of equal score they are sorted chronologically. This produces not just a list of coincidences, but a discrete distribution of the corpus by numerical value.

The essential point is that the pattern must be searched for not only for 68. If one looks only for 68 in advance, one is already in confirmation search mode. It is more correct first to build the distribution of score across the corpus and only then to see which values are singular, which recur and whether the recurrences have a historical structure. Precisely then 68 becomes interesting not because it is a "magic" number, but because in the precisely dated corpus reported here it turns out to be one of the few values with multiple repetitions.

This is the moment when numerology either collapses or becomes a research question. If 68 were simply one of many values with mass repetitions, the topic would quickly close. But when we see that in the compact and verified corpus 49, 54, 67, 85, 97, 107, 127 and 128 appear singly, while 68 appears four times, there is a basis not for superstition, but for careful examination of the cluster itself.

7. What the precisely dated working corpus shows

The working corpus published here includes the following wars and war onsets with clearly verifiable day-month-year dates: the War of 1812, the Crimean War, the Serbo-Bulgarian War, the Italo-Turkish War, the First Balkan War, the Second Balkan War, World War I, World War II, the Six-Day War, the Falkland Islands War, the Persian Gulf War and the full-scale Russian invasion of Ukraine. This is not the total archive of humanity, but it is a sufficiently broad and historically significant layer to show whether 68 is unique or not.

Sorted by score, the corpus yields the following picture: 49, 54, 67, 68, 68, 68, 68, 85, 97, 107, 127, 128. From this follow three observations. First, 68 is not unique, but that is precisely what makes it interesting. Second, in this reported corpus no other score has four repetitions. Third, 68 does not scatter chaotically in time but concentrates on four highly historically charged onsets between 1911 and 2022. If someone searches for, say, "40," in this corpus there is no confirmed case with score 40. There are singular 49, 54, 67, 85, 97, 107, 127 and 128, but there is no other fourfold clustering similar to 68.

An important strict clarification is needed here. This should not be called "outside the normal distribution" in a strict statistical sense, because neither is the corpus a random selection of independent observations, nor is the distribution of historical wars normal. The more correct term is "unusual clustering" or "cluster." Put simply: in the reported set, 68 behaves strangely densely. Stated scientifically: we have an observable repetition of the same value in four cases with high historical weight, but without the right to automatically conclude the existence of a hidden causal law.

8. Chronological order of verified cases with names and brief description

The list is chronological and illustrative of the corpus used, not a claim to be a complete catalog of all precisely dated wars.

18.06.1812 - War of 1812 - score 54. Brief description: a war between the United States and Great Britain, started after an American declaration of war and connected to disputes over maritime rights, trade and imperial rivalry. It is significant for North American history but does not form a 68 pattern.

04.10.1853 - Crimean War - score 85. Brief description: a war between Russia and a coalition around the Ottoman Empire, Great Britain, France and later Sardinia-Piedmont, important for the European balance of power and early modern military logistics.

14.11.1885 - Serbo-Bulgarian War - score 128. Brief description: a short Balkan conflict showing the instability of the post-Berlin order and the significance of the Bulgarian Unification.

08.10.1912 - First Balkan War - score 49. Brief description: a coalition of Balkan states attacks the Ottoman Empire and accelerates the disintegration of the Ottoman order in Europe.

29.06.1913 - Second Balkan War - score 67. Brief description: the former Balkan allies enter a new war over the partition of Macedonia; the score is immediately adjacent to 68 but does not reach it.

05.06.1967 - Six-Day War - score 97. Brief description: a short but strategically enormous Arab-Israeli war, changing the control over key territories in the Middle East.

02.04.1982 - Falkland Islands War - score 107. Brief description: a short, sharp interstate war between Argentina and the United Kingdom over island sovereignty.

16.01.1991 - Persian Gulf War - score 127. Brief description: an international coalition launches an operation against Iraq after the occupation of Kuwait, demonstrating the new post-Cold War military regime.

This chronological order shows two things simultaneously: the majority of verified cases do not yield 68, but when 68 does appear, it appears upon historically exceptionally charged onsets. This is precisely what demands greater attention to the four cases, without allowing the analysis to yield to the false convenience of turning them into "proof" of something beyond the observable.

9. The first 68 case - the Italo-Turkish War as an underestimated marker

The Italo-Turkish War often stands on the periphery of mass historical memory because its immediate scale is limited compared to the subsequent catastrophes. But this is exactly why it is so important for the present analysis. Here we see 68 not upon an already universally recognized world catastrophe, but upon a conflict that appears smaller yet triggers processes with a considerably broader historical radius. Italy undertakes the aggression against the Ottoman Empire in pursuit of colonial expansion in North Africa, attacking the Ottoman provinces of Tripolitania and Cyrenaica, present-day Libya. Formally the war lasts from 29.09.1911 to 18.10.1912, approximately 385 days.

This conflict has at least three levels of significance. First, it reveals and demonstratively exposes the weakness of the Ottoman Empire at a moment when the European system is already under strain. Second, it unleashes and legitimizes a more aggressive national-expansionist line in Italy. Third, this is precisely where the first use of an airplane in war and the first aerial bombing from an airplane occur, giving the conflict the technological status of a prologue to a new type of modern warfare. In other words, this date does not stand upon some peripheral exoticism. It stands upon the early unlocking of the twentieth century as a century of industrialized, technologically accelerated and politically overflowing war. This is exactly why the small direct scale should not obscure the large historical radius.

Regarding casualties and cost, one must speak carefully here. The available figures are more uncertain and more dispersed than for World War I and World War II, and different secondary sources vary. The safest claim is not the absolute number but the structural fact: the direct human devastation is considerably smaller than in the later world wars, but the strategic impact is disproportionately large. This is precisely what makes the Italo-Turkish War a special type of marker. It is not gigantic in number of participants and casualties, but it is large as a revelation that the system can unravel much faster than old diplomacy is ready to admit.

10. After 1911 comes an immediate chain - the Balkan wars and the logic of acceleration

If one searches for whether after the first 68 case there indeed follows a "chain reaction," the closest and most sober answer is precisely to trace 1911-1914. The Italo-Turkish War is not alone. It is followed almost immediately by the First Balkan War in 1912 and the Second Balkan War in 1913. These two wars do not yield 68. The first gives 49, the second 67. But that is precisely the interesting point: in numerical terms, 68 does not spill indiscriminately across all neighboring conflicts but remains concentrated. Historically, however, after 29.09.1911, a short and dense sequence of wars indeed opens, shaking the foundations of the Ottoman and European order.

The First Balkan War begins on 08.10.1912, when Montenegro opens military operations against the Ottoman Empire, and the other allies soon join. It ends with heavy territorial losses for the Ottoman Empire in Europe. The Second Balkan War begins on 29.06.1913, when Bulgaria attacks Serbian and Greek forces. The very fact that the score here is 67 - immediately adjacent to 68 - is interesting only as a curiosity, not as proof. The historical meaning is different: the region enters a phase of accelerated instability, and the system of alliances and fears thickens critically on the very eve of 1914.

Thus the first 68 case does not stand isolated. It precedes a dense corridor of wars in which regional aggression, imperial weakening, nationalism, revisionism and unfinished territorial disputes begin to combine into a larger European crisis. This does not mean that the number 68 "causes" the Balkan wars. It means that if one wants to test the idea of a marker, the Italo-Turkish War is a far more suitable candidate for a marker of unlocking than for a self-sufficiently closed episode.

11. The second 68 case - World War I as a systemic collapse, not just a war

World War I begins on 28.07.1914, when Austria-Hungary declares war on Serbia. It is important here to keep the exact start date separate from the causal prologue. The assassination in Sarajevo is on 28.06.1914, but that is not the date on which the pattern is calculated. Score 68 arises not from the assassination of Franz Ferdinand, but from the moment when the local crisis transitions into a formal military act and the cascade of mobilizations and declarations of war begins. In this sense, 28.07.1914 is not simply a date of onset but a date of systemic switching.

The significance of World War I is of an order that changes the very understanding of scale. This is no longer a regional conflict with possible secondary repercussions, but an international conflict that drew in most of the states of Europe, Russia, the United States, the Middle East and other regions. The war lasts until 11.11.1918, approximately 1,567 days. Britannica's estimates point to about 8.5 million soldiers killed and approximately 13 million civilian casualties, with tens of millions wounded, displaced or permanently affected. Four imperial dynasties collapse. Consequently, here 68 stands upon an event that does not simply open a new conflict but disintegrates an entire historical architecture of power and legitimacy.

From the perspective of the present thesis, World War I is the decisive case because it legitimizes the idea that the 68 cluster is not composed of arbitrary and equally minor incidents. On the contrary, the second 68 case is already an outright civilizational rupture. If the first 68 was a marker of unlocking, the second is a marker of systemic collapse. And here begins the next line of the argument: whether after 1914, 68 reappears only as a numerical coincidence, or again stands upon an event that does not close the tension but carries it forward in an even more destructive form.

Even more important is that World War I cannot be thought of only as a completed historical block between 1914 and 1918. It is the matrix of the twentieth century. It industrializes mass killing, turns total mobilization into a normal instrument of the modern state, accelerates the technology of mass destruction, creates new regimes of propaganda and disintegrates the old imperial frameworks without producing a new stable order. In this sense, the start date 28.07.1914 does not mark simply the beginning of a war, but the beginning of a new historical environment in which war turns out to be no longer an exception but a central mechanism of systemic rearrangement.

If one searches in the 68 cluster not for magic but for a structural hint, World War I is the place where this hint becomes strongest. Because here we are no longer talking about a limited colonial aggression, but about the European system itself transitioning into a mode of self-destruction. Therefore, from the perspective of the present thesis, 28.07.1914 stands not simply as the second numerical case, but as the point at which the pattern touches the very threshold between the old European balance and the modern epoch of cascading global catastrophes.

12. Between 1914 and 1939 - how the second 68 prepares the third

The historical bridge between World War I and World War II is perhaps the strongest argument that with these dates one should look not only at the date itself, but also at the subsequent structure born after it. World War I ends, but it does not stably settle the questions of security, revenge, national humiliation, economic collapse, new borders and collapsed empires. The Versailles order is simultaneously punitive, incomplete and unsustainable. In this sense, 1914 is not only the beginning of a war but the beginning of a long interval of unstable interregnum.

It is precisely here that the numerical pattern begins to acquire a particular interpretive density. If 1911 precedes a short regional corridor to 1914, then 1914 opens a longer but no less dramatic corridor to 1939. This is not a "magical sequence," but a sequence of historically unclosed conflicts. The economic destructions, the debt architecture, the trauma of mass mobilization, the radicalization of politics and the legitimation of extreme regimes accumulate during the interwar period to a point where a new pan-European catastrophe becomes possible and then actual.

It is particularly important that the interwar period is not a passive pause between two ready-made catastrophes. It is a laboratory of the unfinished collapse. In it, new doctrines of mass mobilization are born, new regimes of total ideologization, new claims to state omnipotence and new forms of economic desperation. In other words, World War I does not only cause defeats. It produces the structural conditions in which the next, even more destructive war becomes historically conceivable and politically executable.

Therefore, if we use the language of "markers" at all, we must be very precise. The marker is not the cause. The marker is the threshold after which the system begins to rearrange its risk in a new way. In this sense, 28.07.1914 is not just a date of war, but a date of a new regime of historical instability, which in less than a quarter century produces the third 68 case.

13. The third 68 case - World War II as the globalization of catastrophe

World War II begins on 01.09.1939, when Nazi Germany attacks Poland. Here the moral and causal picture are clearer than in many other cases. This is an act of aggression, standing in a continuum with earlier German expansion and the failure of the policy of appeasement. The pattern again does not "justify" anything and does not soften the responsibility. On the contrary, it places upon an absolutely unambiguous aggressive start the same number that we have already seen in 1911 and 1914.

If World War I is a systemic collapse, World War II is the globalization of the collapse logic. This is no longer just a conflict of empires and alliances, but a total conflict of continents, industries, ideologies, resources and civilian populations. Here war becomes a fully planetary process in which the front and the rear cease to be clearly distinguishable. Cities, transport networks, energy systems, population, workforce and memory become part of the battlefield itself. Mass destruction is no longer a side effect but a systemic mode of operation.

In terms of scope, duration, casualties and cost, this third 68 case is incomparably heavier than the previous ones. The war lasts in its European starting sector from 01.09.1939 to 1945 and becomes the most destructive conflict in the history of humanity. Tens of millions of people perish, vast parts of Europe and Asia are devastated, and the material costs exceed the imagination of previous generations. But even this formula is insufficient, because the true cost of World War II is not only in the dead and the destroyed property, but in the fact that it opens the nuclear age, the Cold War order and the permanent possibility of civilizational self-destruction.

From the perspective of the cluster, this is the decisive moment. If 68 had repeated at 1911 and 1914 and then dissolved into dozens of similar repetitions upon secondary wars, the thesis of its distinctiveness would have weakened. But when the third repetition occurs on 01.09.1939, the number begins to mark not just wars but start dates of conflicts that radically rearrange the world order. This is still not proof of a "cosmic code," but it is already sufficient to speak of a historically strange cluster, not just an internet meme.

14. Why the other late and well-dated wars do not confirm the same pattern

One of the most important arguments in favor of serious verification is precisely the fact that many important wars do not fit into 68. If the pattern were too easy to achieve in the working corpus, it would have no particular value as an observation. But we see that the War of 1812 gives 54, the Crimean War 85, the Six-Day War 97, the Falkland Islands War 107, the Persian Gulf War 127, and the Serbo-Bulgarian War 128. That is, historical significance by itself does not lead to 68. There are important, even pivotal wars that fall on entirely different values.

This fact is extremely important because it protects the analysis from a subtle but dangerous illusion. If someone sees 68 and decides that "all great wars" yield 68, they are already in a mode of false totalization. The reality is different. The pattern is not general for all great wars. It is selective. It is precisely this selectivity that makes the question more interesting but also more difficult. Because it means that we cannot reduce the result to either a banal coincidence or a trivial universal law. We have a more uncomfortable case: a partial, concentrated, historically loaded repeating motif.

Here something else must be added. Some of these "non-matching" wars have enormous weight in their own regions and historical systems. The Crimean War changes the European equilibrium and exposes the logistical weaknesses of modern armies. The Six-Day War has a colossal impact on the Middle East. The Persian Gulf War demonstrates a new regime of high-technology coalition warfare. If the pattern were simply a function of the magnitude of the event, it should appear much more frequently. Precisely because it does not appear, the 68 cluster remains analytically uncomfortable and therefore interesting.

From this follows a more general conclusion. In historical analysis, absences are as important as coincidences. The fact that many visible wars do not yield 68 stabilizes the very observation about the four cases. They are not interesting because everything is 68. They are interesting precisely because not everything is 68, yet despite this, 68 returns at several exceptionally loaded points.

15. The fourth 68 case - 24.02.2022 and the problem of the phases of the war in Ukraine

With Ukraine, methodological discipline is mandatory because here there are two different beginnings. The broader Russo-Ukrainian war begins in February 2014, when Russia occupies Crimea and supports the war in Donbas. But precisely what the internet pattern uses and what the present article examines is not the general onset of the conflict, but the beginning of the full-scale Russian invasion on 24.02.2022. This distinction is not cosmetic. It is the core of honest calculation. If one silences the phase problem, the score becomes rhetoric, not analysis.

The events after 24.02.2022 are sufficiently massive to justify the attention to this fourth case even without the numerical pattern. Russia undertakes an open and large-scale aggression against a sovereign state. The attempt to quickly break Ukraine fails, and the war turns into a long conflict of attrition, industrial capacity, mobilization, drone and missile adaptation, Western aid, sanctions and geopolitical rearrangement. The very fact that this is a European war involving a nuclear power and deep involvement of external support systems makes its escalation potential structurally higher than that of most post-war conflicts. To this is now added the fact that the war acts as an open proving ground for AI, autonomous systems and data-driven combat cycles.

In terms of human and material cost, it is reasonable here to speak in the categories of an unfinished conflict. Unlike the three older 68 cases, this one is not yet historically closed. Therefore, any precise final figure would be temporary. But precisely this amplifies its significance. This is not about an archived catastrophe, but about an open historical system that continues to produce casualties, destruction, displacement, budgetary exhaustion, industrial readjustment and geopolitical rearrangement. Here the numerical marker stands upon an event whose ultimate weight cannot yet be fully calculated.

This makes 24.02.2022 a particularly important date within the framework of this article. If 1911 is an early unlocking, 1914 a systemic collapse and 1939 a global catastrophe, then 2022 is an open point of high historical uncertainty. It has already rearranged European security, broken old economic and energy dependencies, changed the views on the industrial base of war and opened a new epoch of mass use of drones, sensors and digitally assisted combat management. In this sense, the fourth 68 case is not simply a repetition of the number, but a new threshold of war under conditions of mutual technological, infrastructural and nuclear vulnerability.

16. Whether the four 68 cases truly look like markers

If we use completely sober language, the most correct formulation is the following: in the reported precisely dated corpus, score 68 forms a historically unusual clustering upon four conflicts or war onsets with high systemic significance, located in the interval 1911-2022. This is already a sufficiently strong claim and does not need to be embellished. But if we take one step toward interpretation, we see something else: these four cases are not simply large wars. They are wars after which the historical field is rearranged and gives birth to new chains of conflicts, institutions, technologies and forms of destruction.

The Italo-Turkish War precedes the Balkan explosion and the accelerated exposure of Ottoman weakness. World War I unlocks an interwar order that does not stabilize Europe but carries it toward revenge and radicalization. World War II creates the Cold War structure, the nuclear threshold and the post-war international architecture. The full-scale invasion of Ukraine has already rearranged European security, energy dependencies, the military industry and the attitude toward sovereignty and borders. In this sense, "marker" does not mean a mystical sign. It means a beginning after which a series of secondary and tertiary historical rearrangements starts.

It is precisely here that the subtlest danger also appears. The marker language easily slips toward teleology, as if history were a written script and the number merely shows us the page. The present text rejects such a reading. It is more correct to say that the four 68 dates have a retrospective marker function in our description because they stand upon starting points of strongly unfolding historical chains. Who "marked" them, if it is at all appropriate to use such a verb, we do not know. The scientific language for now only allows us to say that the cluster is genuinely observable, while the reason for the clustering remains unproven.

And yet there is something here that must not be trivialized. Four 68 cases in a relatively short modern horizon, given that a number of other wars do not fall upon this value, create a sense not of proof but of historical asymmetry. Some numbers behave as background. Others - at least in this corpus - begin to behave as local concentrators of attention. This is sufficient to justify examining 68 not as truth, but as an anomaly that should be kept in the archive of serious but unproven observations.

17. What this pattern does not prove and how it should not be used

First, the pattern does not prove that there exists an external intelligence, a cosmic code, a secret calendar of wars or a pre-written script of human history. Second, it does not prove that numbers govern politics. Third, it does not prove that a war is "predetermined" simply because its start date yields the same score as another. Fourth, it should not be used to justify aggressors, to relativize guilt or to turn human violence into an aesthetic puzzle. If 68 stands upon the start date of aggression, this does not soften the aggression. It merely adds a numerical layer to the archive of the event.

The scientific minimum here requires thinking in several directions simultaneously. It is possible that 68 is simply a rare but entirely random repetition upon a small and historically selected corpus. It is possible that there is a selection effect - that people notice precisely these cases because they are world-famous while neglecting the many other conflicts with different values. It is possible that the format DD + MM + CC + YY is flexible enough to produce impressive but post-hoc selected coincidences. It is also possible that the cluster is real but is only a statistical oddity without deeper historical meaning.

There is also a more specific danger: that the pattern replaces the analysis of causality. Wars do not start because of numbers. They start because of decisions, interests, ideologies, fears, ambitions, revisionism, imperial inertias, collapses of deterrence and failures of the political order. If one forgets this and lets the number take the place of the historical mechanism, one does not merely err. One misses the very structure of responsibility. In this sense, the number can be observed, but it must never be placed in the place of history.

Nevertheless, it would be entirely wrong to do the opposite as well - to dismiss the pattern with a contemptuous wave of the hand just because it sounds numerological. True discipline is neither credulity nor cynicism. It is the ability to hold the strange long enough to check what exactly is strange about it. In this case, the "strange" is not the number itself. What is strange is that four precisely dated war onsets with enormous historical density turn out to be upon the same score, while many other important wars in the corpus do not line up in this way.

18. The radical hypotheses - permitted only as clearly labeled speculation

The following assumptions are speculative and do not have the status of established knowledge.

Once the methodology has been described, the limitations set and the observation honestly formulated, the last and most risky layer can be opened: the radical assumptions. If one steps outside conventional historical thinking, several extreme readings of the 68 cluster are possible.

The first is cosmological-stochastic: the Universe allows local symmetries and clusterings that have no agent, only a strong symbolic effect for the observer.

The second is informational-archival: the human calendars themselves and ways of remembering beginnings may produce patterns that are not "in the world" but in the interface between the world and memory.

The third reading is more radical and more absurd by current standards: that certain historical events leave not only causal but also formal traces in the way they arrange themselves in time, as if history had a faint signature of its own.

The fourth, even more speculative, is the signal hypothesis: that certain dates function as messages not from an external intelligence but from the future itself as a retrospective structure - that is, events with the greatest subsequent branching turn out to be "visible" in more than one dimension of the archive.

The fifth is the conspiracy version in the strict sense - that some centers of power deliberately choose dates, knowing about symbolic regimes. This last hypothesis is weak because it requires excessively strong assumptions and has no reliable evidence.

A sixth, even more extreme speculation can be added, which by current scientific standards remains entirely unprovable but is philosophically illustrative: that history is not only a linear series of past events but a structure in which future states can retrospectively "condense" the significance of certain starting points. In such a reading, 68 would look not like a code embedded in advance, but as a faint resonance between events that, from a later perspective, turn out to be connected by a common ability to unlock cascading epochal transformations. This sounds absurd by current understandings, but it is useful as a thought experiment because it shows us how quickly the human mind tries to fill the gap between observation and explanation.

From the position of scientific rigor, none of these radical hypotheses can be accepted as established. But from the position of "History of the Future," it is useful to voice them, as long as they are properly labeled. They are not knowledge. They are boundary thought modes that show how far human interpretation is tempted to go when it encounters an unusual coincidence with a large historical charge. The most mature conclusion is not "this is a sign" nor "this is nonsense." The most mature conclusion is harder: the observation is real, the cluster is strange, the cause remains unestablished, and the task of analysis is to flee neither toward mysticism nor toward lazy denial.

19. The fourth 68 case as an open corridor toward systemic rearrangement - scenarios for continuation

If 1911 is an early unlocking, 1914 a systemic collapse, and 1939 a global catastrophe, then 2022 does not look like a closed war with a limited theater, but like an open corridor of historical rearrangement whose secondary and tertiary consequences have not yet fully unfolded. It is precisely here that the fourth 68 case begins to differ from the previous ones not by being less important, but by the fact that we are still inside it and therefore see not its entire historical silhouette, but only its active unfolding frontline. It has already rearranged European security, broken previous energy dependencies, accelerated the remilitarization of the industrial base, returned the question of reserves, production capacity and the pace of armament to the center of state strategy and opened a new epoch of mass use of drones, sensors, electronic warfare, space-assisted targeting chains and digitally coordinated combat management.

An important clarification - in the present article, the word "theater" is used as an analytical term - for a space of conflict, zone of war and risk, field of violence and escalation or a connected zone of crisis - but not as a language of distance, indifference or depersonalization. Behind every such "theater" stand real people, real bodies, real families, destroyed homes, interrupted biographies, grief, fear, hunger, exhaustion and long human consequences that cannot be reduced to a map, an index, or a strategic scheme. The victims are not merely numbers. They are human lives, human pain and human loss. If this text works with structures, thresholds, accelerators and connected spaces of conflict, it is not to turn suffering into an abstraction, but to make more visible the mechanisms that produce real death, real destruction and the real long wound in people's lives.

If viewed strictly through the logic of the first three 68 cases, what they have in common is not simply the size of the war, but the fact that each subsequent one stands upon a higher level of systemic risk and a broader field of subsequent fractures. The Italo-Turkish War unlocks a regional corridor toward the Balkans and 1914. World War I unlocks the collapse of empires, inter-systemic radicalization and a corridor to 1939. World War II unlocks the nuclear age, the Cold War order and the permanent possibility of civilizational self-destruction. If this ascending order of fatality is not accidental but at least historically describable, then 2022 must be thought of not simply as a war in Europe, but as the beginning of a new phase in which the risk is no longer only territorial or regional, but simultaneously military, infrastructural, energy-related, cybernetic, financial, logistical and alliance-related.

This is precisely why the fourth 68 case should be viewed as a possible threshold of a multisectoral cascade. In the previous three cases, the consequences did not remain locked within the initial theater. They transferred into diplomatic regimes, military doctrines, new alliances, new technologies and new types of vulnerability. By analogy, 2022 can be read not as a single conflict, but as the start of an epoch in which several crises, formally distinct from one another, begin to feed each other. This does not mean that a single war automatically "produces" all subsequent ones. It means that the fourth 68 case may turn out to be the start date of a new regime of connected crises, in which every local tension is amplified by the already shaken general architecture of order.

The first realistic scenario for continuation is a scenario of long erosion and disintegration of the stabilizing frameworks, without necessarily having a one-time theatrical "explosion" that everyone immediately calls World War III. This would be not a classic world war of the 1914 or 1939 model, but a polycentric epoch of layering: the war in Ukraine continues to exhaust Europe and Russia; the Middle Eastern escalation shakes energy, shipping and prices; the allied relationships between the United States and European states enter a period of doubt and conditionality; peripheral territorial disputes begin to be read no longer as local cases but as tests of will, response and credibility of guarantees. This is a scenario not of instant apocalypse, but of structural dissolution of order.

The second scenario is one of torn alliance architecture. In it, the fourth 68 case does not immediately lead to a unified global front, but to something more dangerous in the long horizon - divergence in the strategic centers of the West. If within NATO and the European Union the doubt deepens whether the United States remains a reliable, predictable and strategically consistent guarantor, European states will begin to move not as a unified system, but as a collection of semi-coordinated reactions. Some will accelerate militarization, others will seek a deal, still others will try a dual course, and others will fall into political paralysis between internal fatigue and external risk. This would create a dangerous gap in which adversaries test boundaries while allies have not yet decided what exactly they are ready to defend and at what cost.

The third scenario is one of horizontal spillover to new theaters without formal unification of the conflicts. This is a particularly likely type of development in an epoch in which crises of different origins begin to be perceived as elements of one general global transition. If the Middle Eastern war continues to maintain an open energy and cybernetic front, if the Venezuela-Guyana dispute remains an active territorial and resource dispute with escalation potential, if the tensions around Greenland and the broader Arctic sensitivity continue to shake trust within the alliance architecture, and if the Ukrainian front continues to demand enormous financial, industrial and political resources, then the world will enter a regime in which the separate crises do not merge juridically but begin to amplify each other strategically. This is precisely one of the most dangerous variants, because it makes the global crisis real without giving political elites the convenient name under which societies would recognize it in time.

The fourth scenario is one of accelerated nuclear and quasi-nuclear nervousness. It is not necessary for a nuclear weapon to be used for the system to begin functioning as nuclear-stressed. It is sufficient for deterrence to become unclear, red lines to blur, external actors to move closer to direct confrontation, and the chains of warning, intelligence and command to operate under constant stress. Then every regional escalation begins to have disproportionate significance. A strike on infrastructure, a misread deployment, an attack through a proxy network or a major cyber incident can produce a chain of decisions taken not in strategic mode but in emergency mode. If the fourth 68 case has higher fatality than the previous ones, it may manifest precisely this way - not as a symbolic "last war," but as an epoch in which the risk of systemic error becomes a permanent background, and the very maintenance of strategic sobriety becomes a hard-to-achieve political resource. In such a world, the danger comes not only from malicious action but also from the accumulation of tension in a system that gradually loses its ability to distinguish a limited signal from an existential threat.

The fifth scenario is one of internal disintegration of democratic societies under the pressure of external wars. This is often underestimated because the mass imagination looks for the map of the fronts rather than the map of fatigue. But each of the previous large 68 points produced not only external but also internal transformation - mobilization, authoritarianization, control, propaganda, economic discipline, new forms of coercion. By analogy, 2022 may turn out to be the beginning not only of a series of external conflicts, but also of an internal mutation of Western societies: higher tolerance for emergency regimes, political radicalization, cultural division, erosion of trust in institutions and redistribution of resources from the social to the defense state.

This scenario is particularly important because it shows how the fourth 68 case may prove fatal not only through the front, but through the corrosion of the very societies trying to withstand the external pressure. When fear, economic burden, migration pressure, information warfare and the sense of endless crisis begin to accumulate, political systems no longer think long-term but reactively. In such an environment, the demand for strong solutions, quick sanctions, emergency powers, simplified enemies and crude political oppositions grows. This is not a secondary effect. It is part of the very mechanism by which a war can escalate into a much broader civilizational rupture.

If the first three 68 cases show that great wars do not simply destroy territories but produce new regimes of internal order, then 2022 must be thought through the prism of the political psychology of exhaustion. Prolonged war does not only kill people and infrastructure. It changes the standard for normalcy. It makes societies accustomed to peace adapt to permanent anxiety, scarcity, threat and tension. This is precisely one of the possible deepest consequences of the fourth 68 case - not just new fronts, but a new type of political population living in a permanent mobilization environment.

The sixth scenario is the heaviest and must neither be silenced nor presented as fact. This is the scenario in which the fourth 68 case truly turns out to be a prelude to something that future generations would call World War III, even though contemporaries long refuse this name. This scenario does not presuppose a single day of universal mobilization and a general declaration of war. More likely it would develop as a gradual merger of several large crises into one common system of mutual amplification: Ukraine, Iran and its regional orbit (not as a juridically merged part of the Ukrainian war, but as a strategically coupled theater in a broader crisis regime), cyberattacks on critical infrastructure, energy shocks, Arctic tensions, resource territorial disputes and finally the collapse of trust between old allies.

This is precisely what makes the present moment so difficult to read. Historical catastrophes are often misrecognized while they are unfolding. People expect a world war to look like a textbook world war, with a clear beginning, a clear division into blocs and a clear general mobilization. But it is entirely possible that the fourth 68 case leads to a newer form of global war - distributed, non-simultaneous, uneven, but real in its consequences. In such a world, the formal absence of a unified name does not mean the absence of a unified crisis. On the contrary, it may be part of the very danger, because societies and elites are late to understand the scale of what is happening.

It must be clearly underscored here: this is a scenario, not a diagnosis. It is not an established fact that the world is already in World War III. But it is an establishable fact that the fourth 68 case is developing under conditions of much higher mutual connectivity, nuclear nervousness, technological dependence and systemic vulnerability than the previous three. Therefore, if there is a moment when scenario thinking must be bold, it is precisely here - not to predict apocalypse, but to not miss the signs that an open regional war may turn out to be only the first visible layer of a larger planetary rearrangement.

But there is also a seventh scenario, which is less theatrical and may turn out to be even more likely: total rearrangement without a formal world war. In it, the fourth 68 case acts as a catalyst of a new multipolar architecture in which the old post-war and post-Cold War system of guarantees, dependencies and legitimacies definitively erodes. Europe arms itself, but later and more nervously than necessary. The United States acts more selectively and more transactionally. Russia remains a destructive factor even when weakened. China gains space not necessarily through direct war but through waiting, industrial mass and systemic patience. Middle powers begin to act more independently.

This scenario is no less historically heavy just because it does not contain one big name. In fact, it may turn out to be even more deeply transformative, because it does not end with a peace treaty that rearranges the world along a new formula, but with a long period of erosion in which the old order dies faster than the new one is born. In such a development, 2022 will not be remembered only as the beginning of one war, but as the date after which the global system ceased to be convinced of its own architecture. This would be a kind of historical equivalent of systemic decoherence - not a sudden collapse everywhere, but a gradual disintegration of the common framework that had hitherto kept the different theaters, alliances and economic zones in a relatively coordinated state.

In this sense, the fourth 68 case may turn out to be the start date of a new epoch of great strategic loneliness. States will have more weapons, more data and more technologies, but less trust, less stable guarantees and less certainty about which rules still apply. This is a particularly dangerous combination. High connectivity without high legitimacy does not stabilize the world. It only accelerates the transfer of shocks between its separate parts.

If one must speak most boldly but also most honestly: the fourth 68 case already resembles the start date of an epoch, not merely the start date of a war. Its chain reaction is already visible in the shaking of alliances, in the readjustment of energy and military systems, in the growth of cybernetic risk, in the weakening of taboos against direct force, in the return of resource geopolitics and in the sense that the center of the world order no longer holds its peripheries as it did before. This is still not proof that the worst scenario is necessarily forthcoming. But it is sufficient grounds to think that if the first three 68 cases turned out later to be bigger than they seemed at the beginning, the fourth almost certainly has not yet shown its full size.

It is precisely here that both boldness and restraint must be held simultaneously. The boldness is to acknowledge that there is a visible pattern of rising fatality: 1911 unlocks, 1914 collapses, 1939 globalizes the catastrophe, and 2022 already bears the marks of a conflict capable of spilling beyond its initial form. The restraint is not to turn this line into a false prophecy. Historical analysis does not predict fate with precision. It outlines fields of increasing probability, corridors of risk and boundaries beyond which the cost of error becomes civilizationally unbearable.

Therefore, the strictest conclusion is not that "68 predicts" the future, but that when this pattern appears upon wars with such historical density, it is reasonable to treat it as a warning archival signal. Not as a proven law. Not as a mystical code. But as an anomaly that says the following: some start dates do not only mark the beginning of hostilities but turn out to be thresholds after which the world begins to behave in a new way. If this holds here too, then 24.02.2022 will not remain simply a date from one war. It will remain the starting point of a new historical configuration that will either escalate into a much larger common catastrophe, or produce such a deep rearrangement of order that the very phrase "post-war world" will already look like the name of a lost epoch.

20. The weapon as accelerator - the hypothesis that each 68 case opens a new technological regime of war

One of the strongest working hypotheses that can be derived from the four 68 cases is that they mark not just large wars, but thresholds at which a new accelerator enters war - a new type of weapon, a new infrastructure of killing or a new regime of linking reconnaissance, decision and strike. In its strictest form, this hypothesis should not be formulated as the superstitious "every 68 brings a new weapon," but as the more precise claim that every 68 threshold coincides with the entry or consolidation of a new type of military accelerator that expands the scope, tempo and depth of the conflict. If this is true, then the number 68 would not stand upon arbitrary wars, but upon moments in which the very nature of war takes a step forward toward a higher density of destruction.

In the first 68 case - the Italo-Turkish War of 1911 - the strongest candidate for such an accelerator is the aerial dimension of warfare. It is precisely here that the first combat uses of an airplane for reconnaissance and the first aerial bombardments appear, carried out by Italy over Ottoman positions in Libya. This is a historically decisive moment not because the immediate destructive power of these early airplanes is enormous, but because for the first time the battlefield ceases to be entirely two-dimensional. From this moment on, war begins to have a new altitude, and with it a new logic of seeing, correction and strike. The airplane in 1911 is not yet a mass platform for destruction, but it is a prototype of the future regime in which the speed of observation and the speed of strike begin to merge.

This makes the first 68 case particularly important for the present hypothesis. It does not bring the most lethal weapon but the first leap in the position of the observer. War is no longer simply the movement of armies across a map. It begins to become a system in which the one who sees from above changes the very structure of vulnerability. Here the weapon is not just the airplane as a machine. The weapon is the new angle of the world. From this moment on, reconnaissance, fire correction and later bombing will become one of the main lines of modern warfare. In other words, the first 68 does not simply introduce a new technique. It introduces a new geometry of conflict.

In the second 68 case - World War I - the hypothesis must be formulated more broadly. Here it is difficult to point to a single weapon, because the more accurate description is that 1914 introduces a package of accelerators which together turn war into a scientific-industrial process of mass attrition. In this war, chemical weapons, the tank, combat aviation, mass machine-gun fire dominance, improved long-range artillery, radio-communications support and the logic of total mobilization all debut or are consolidated. If 1911 opens the sky, then 1914 makes the entire society a weapons system. This is precisely why the second 68 case must be read not only as the start of a war, but as the start of a new regime in which the weapon is no longer a separate object, but an entire ensemble of industry, chemistry, mechanics, transport, communications and state capacity.

Two lines are particularly illustrative here. The first is chemical warfare - the moment when science begins to enter not only the construction of weapons but the very composition of the air as a combat environment. The second is the tank - the attempt to break the trench stalemate through mechanized protection and movement. Neither of these weapons is fully mature at the start of the war, but precisely this confirms the hypothesis: the 68 cases do not necessarily coincide with a fully finished weapon, but with the threshold at which the new accelerator enters history and begins to transform the battlefield. Therefore the second 68 does not bring a single atomic innovation, but the first great fusion of industry and war in a regime of mechanized mass destruction.

In the third 68 case - World War II - the hypothesis acquires its clearest form. Here one can already point to a weapon that does not simply accelerate the war but opens a new era for the entire civilization: the atomic bomb and the subsequent regime of nuclear warheads. Strictly speaking, the atomic bomb does not exist as a combat-used weapon on 01.09.1939. But this does not refute the thesis - it makes it stronger. The third 68 marks the war within which the nuclear project is born, completed and enters history as a force actually used. If in 1911 the airplane is still weak and in 1914 the tank and chemical weapons are still in an early phase, then in 1939 the nuclear weapon appears as an ultimately finished accelerator that does not merely increase destruction, but changes the very concept of the limit of destructiveness.

This is precisely what makes the third 68 different by an order of magnitude. The previous accelerators expand the front, increase the tempo and densify death. The nuclear weapon changes the very ontology of war, because from this moment on humanity lives in the shadow of the ability to destroy entire cities in seconds, and in its ultimate form - its own civilization. In this sense, the third 68 is not simply another stage in the evolution of weapons. It is the threshold at which the weapon becomes civilizational, not just military. The nuclear warhead is an accelerator not just of victory or defeat, but of the very possibility for politics to fail irreversibly.

In the fourth 68 case - 24.02.2022 - the most serious candidate for a new accelerator is artificial intelligence (AI), but it must be understood correctly. Not as a standalone "magic machine," but as a layer that begins to link reconnaissance, logistics, autonomy, navigation, target detection, strike prioritization and platform adaptation in much shorter cycles than before. The war in Ukraine has already shown the mass use of drones as a primary combat environment, but in 2025-2026 a clearer transition toward AI-assisted and AI-trained systems is also visible: Ukraine opened real combat data to allies and companies for training AI models for drone technologies, speaks of autonomous systems and already acts as a laboratory for the next generation of combat autonomy. Here the accelerator is not just the drone, but the fusion of drone, model, data, jamming environment and fast retraining cycle.

This is a crucial difference. With the atomic bomb, the new weapon is concentrated in a single ultimate explosive power. With AI, the new weapon is not necessarily a single object. It is a networked regime of combat acceleration. It makes possible faster reading of images, faster detection of patterns, faster targeting of autonomous or semi-autonomous platforms, faster transfer of knowledge from one theater to another. In this sense, the fourth 68 case may turn out to be not just a "drone war," but the beginning of an epoch in which combat advantage will increasingly depend not on the individual platform but on who better turns data into decision, decision into targeting and targeting into a mass-reproducible combat practice.

It is important here to make a more precise but also bolder distinction. For Ukraine, there is sufficiently clear and reliable data that the war acts as a real laboratory for AI-drone technologies, autonomous systems and data-driven transformation of the battlefield. For Iran, we are also no longer speaking purely hypothetically. The publicly described role of the Palantir Maven Smart System, in which Anthropic's Claude was integrated, shows a real case of AI-assisted targeting workflow: the system extracted meaning from large volumes of classified data, proposed hundreds of targets, provided precise coordinates and prioritized targets by importance, accelerating planning from weeks to near real-time. This is already qualitatively different from the earlier, more cautious formulation about a "general AI regime." Here we have a publicly described real military use of AI as a kill chain accelerator.

But precisely here the boundary of honest claim must remain. From these data it does not follow that Anthropic or Claude independently "directed all 6,000+ targets" in the sense that the sensational headline suggests. The more correct claim is different: Claude, integrated in Maven, was part of a real targeting and prioritization workflow in the American campaign against Iran, with a visible role in accelerating target generation, location assignment, prioritization and post-strike evaluation. This is a sufficiently strong historical claim and does not need to be dramatized beyond what is available.

Regarding the Maduro case too, it is now more accurate to speak carefully but not excessively timidly. Public data indicate that Claude was used in the operation to capture him as well, but the reliable public picture there is weaker and less unraveled than with Iran. The case is real and illustrative as an expansion of the operational environment, but remains a weaker evidentiary anchor case for AI-centrality than Iran and Ukraine. Therefore the Maduro case can stand as an indicator of expanding operational use of the same class of instruments, but not as the main evidentiary anchor point for the theory.

If we bring the four cases together in one line, the hypothesis acquires an even stronger form: the first 68 introduces the aerial view and the aerial strike; the second 68 introduces the scientific-industrial package of mechanized mass destruction; the third 68 introduces the nuclear limit; the fourth 68 no longer merely "hints" but publicly demonstrates algorithmic and data-driven warfare, in which the weapon is a system for accelerating the entire kill chain - from detection to decision, from decision to strike, from strike to a new cycle of assessment and learning. In this framework, the weapon is an accelerator not only because it kills more, but because it compresses the time between perception and destruction.

It is precisely here that the hypothesis becomes truly alarming. If with each subsequent 68 case the accelerator becomes less local and more civilizationally encompassing, then AI in the fourth 68 may prove to be an even deeper threshold than the atomic bomb in one specific sense. The nuclear weapon is the ultimate limit of destruction. AI is a possible accelerator of the entire system that decides when, where, how and with what destruction is to be applied. The nuclear warhead concentrates power. Algorithmic war concentrates tempo. And in an era of high connectivity, tempo is often more dangerous than the single platform itself, because error, escalation and mass reproduction of strikes become faster than the political capacity for correction.

21. Weapon Acceleration Theory (WAT) and Weapon Acceleration Index (WAI)

Based on the analytical power of the "History of the Future" framework, which I created and developed as the foundation of Gen FUTURE, for the purposes of the present analysis, I formulate the Weapon Acceleration Theory (WAT) and the Weapon Acceleration Index (WAI).

21.1. Clear formulation of the theory

The Weapon Acceleration Theory (WAT) formulates the following claim: for those 68-cases that are not merely numerical coincidences but events with high systemic significance, their historical weight is partially explained by the entry of a new military accelerator. "Military accelerator" means not just a new weapon as an object, but a new regime that compresses the time between observation, recognition, decision and strike, expands the spatial scope of the conflict and leaves a lasting trace in the next military epoch.

In strict form, the theory claims that the four 68-thresholds form a sequence of ascending accelerators: 1911 - the aerial view and aerial strike; 1914 - the industrial-mechanized package of mass destruction; 1939 - the nuclear limit; 2022 - algorithmic, data-driven and partially autonomous acceleration of the combat cycle. If this line is valid, then the 68-cluster does not mark "magical dates" but thresholds at which war changes its primary carrier of acceleration.

The theory is scientifically useful only if treated as falsifiable. It weakens if it is shown that the 68-cases do not coincide with new accelerators, or if the current case from 2022 does not lead to a lasting reformatting of the combat cycle through AI, autonomous systems and machine-assisted targeting. It strengthens if the current case continues to produce doctrinal, technological and alliance transformation comparable in depth to the effect of aviation, total industrial war and the nuclear regime.

21.2. Operationalization of the theory

To ensure that WAT does not remain merely a qualitative hypothesis, the Weapon Acceleration Index (WAI) is introduced here, normalized from 0 to 100. The index does not measure the "evil" of war but the strength of the new accelerator introduced or consolidated by the corresponding 68-case.

Formula:

WAI_i = N_i + K_i + S_i + P_i

where:

N_i = Novelty of the accelerator, from 0 to 25
K_i = Kill-chain compression, from 0 to 25
S_i = Scale amplification / Increase in scope and strategic depth, from 0 to 25
P_i = Persistence and diffusion / Durability and diffusion in the next military epoch, from 0 to 25

Normalization:

0-25 = weak or local accelerator
26-50 = significant but limited accelerator
51-75 = strong accelerator with long-term consequences
76-90 = very strong accelerator with epochal transformation
91-100 = civilizational accelerator, changing the very framework of war

21.3. Rules for scoring the four components

N_i is scored by the degree to which the new accelerator introduces a new class of combat capability. A value of 25 means not just an improvement, but a new military regime.

K_i is scored by the degree to which the time between detection, interpretation, target selection and strike is shortened. A high value means that tempo becomes an independent source of advantage.

S_i is scored by whether the accelerator remains tactical, becomes operational, strategic or civilizationally significant. Nuclear weapons, for example, have a near-maximum value because they change not only the front but the very limit of political error.

P_i is scored by whether the accelerator remains episodic or becomes a permanent element of the following decades. Here, not the momentary shock but the historical carrying capacity of the new regime is assessed.

21.4. Calculation for the first three 68-cases and current value for the fourth

I) 29.09.1911 - Italo-Turkish War Primary accelerator: aerial reconnaissance and early aerial bombing

N_1911 = 24
K_1911 = 9
S_1911 = 11
P_1911 = 20

WAI_1911 = 24 + 9 + 11 + 20 = 64

Interpretation: the first 68 introduces a new geometry of warfare. It does not yet carry maximum destructiveness, but opens the vertical dimension and makes the aerial view and aerial strike permanent elements of future conflicts.

II) 28.07.1914 - World War I Primary accelerator: industrial-mechanized package - chemical weapons, tank, mass aviation, machine-gun fire dominance, heavy artillery, total mobilization

N_1914 = 19
K_1914 = 20
S_1914 = 21
P_1914 = 22

WAI_1914 = 19 + 20 + 21 + 22 = 82

Interpretation: the second 68 does not introduce a single weapon, but a new regime of war in which industry, science, logistics and mass mobilization begin to act as a combined accelerator of destruction.

III) 01.09.1939 - World War II Primary accelerator: nuclear threshold, realized within the war through the atomic bomb and the subsequent regime of nuclear warheads

N_1939 = 25
K_1939 = 22
S_1939 = 25
P_1939 = 25

WAI_1939 = 25 + 22 + 25 + 25 = 97

Interpretation: the third 68 introduces a civilizational accelerator. The nuclear weapon does not simply increase damage but changes the very concept of the limit of destruction and creates a long post-war regime of nuclear deterrence, fear and systemic vulnerability.

IV) 24.02.2022 - Russia's Full-Scale Invasion of Ukraine Primary accelerator: AI-assisted, data-driven and partially autonomous warfare with mass drone use, accelerated learning cycle and already publicly visible cross-theater diffusion to other military campaigns, including through Palantir Maven + Claude in the American campaign against Iran.

N_2022,current = 23
K_2022,current = 25
S_2022,current = 19
P_2022,current = 23

WAI_2022,current = 23 + 25 + 19 + 23 = 90

Interpretation: the current fourth 68 is already in the zone of a very strong epochal accelerator. What is new here is not only the Ukrainian theater, but the fact that the AI-assisted targeting and prioritization workflow already has publicly visible diffusion in other operations as well, making the fourth 68 less of a local innovation and more the beginning of a new general military regime.

21.5. Why the current value of the fourth 68 is already 90 and not 87

The current case receives maximum K because AI and autonomous systems no longer merely assist peripheral functions but visibly compress the kill chain time through accelerated target generation, prioritization, battle planning and post-strike assessment. It receives slightly elevated N because we now have more clearly proven transition from theatrical innovation to a broader operational workflow. S remains below the nuclear maximum because AI acceleration has not yet reached the existential limit of nuclear weapons. P remains high because diffusion is no longer hypothetical: movement is visible from the Ukrainian theater to other real operations (AI is already not only an accelerator in the Ukrainian theater, but a publicly described operational workflow in Iran as well).

In other words, AI in the fourth 68 is no longer just a "promise" or "expectation." It is a verifiable presence in the real process of military selection, prioritization and acceleration of strike. This does not mean the system is fully autonomous. It means the threshold of algorithmic warfare has been crossed in a practical, not merely theoretical sense.

21.6. Growth model for the fourth 68 index

We preserve the same model of approaching the ceiling of 100:

c_12 = (WAI_1914 - WAI_1911) / (100 - WAI_1911)
c_23 = (WAI_1939 - WAI_1914) / (100 - WAI_1914)

c_12 = (82 - 64) / (100 - 64) = 18 / 36 = 0.50
c_23 = (97 - 82) / (100 - 82) = 15 / 18 = 0.8333

Average coefficient of historical closure toward the ceiling:

c_bar = (0.50 + 0.8333) / 2 = 0.6667

Projection for the mature phase of the fourth 68:

WAI_2022,projected = WAI_2022,current + c_bar x (100 - WAI_2022,current)

WAI_2022,projected = 90 + 0.6667 x 10 = 90 + 6.67 = 96.67

Rounded central projection:

WAI_2022,projected ≈ 97

21.7. Interpretation of the new projection

The central working projection of the theory is now that the fourth 68 is likely moving toward an index of approximately 97, that is, toward a zone in which algorithmic, autonomous and data-driven warfare will acquire systemic weight comparable in epochal depth to the third 68, albeit through a different mechanism. In 1939, the civilizational threshold is the limit of destructive power. In 2022 and beyond, the threshold is the limit of accelerated decision, accelerated targeting and accelerated escalation.

The upper hypothetical boundary of 98-100 remains possible but should not be claimed prematurely. It would only be reached if AI transitions from assisting and accelerating to a nearly universal operational layer of war - with lasting penetration into swarms, autonomous countermeasures, battle planning, theater coordination, infrastructure attack and strategic decision support. At present, the most honest position is that the fourth 68 has already crossed the threshold of demonstrable seriousness, but has not yet fully developed its ultimate historical form.

21.8. The strictest conclusion

WAT does not claim that the number 68 causes new weapons. It claims something more limited and stronger: that in the observed 68-cluster a sequence is visible in which each subsequent war introduces an accelerator at a higher level of abstraction and higher systemic risk - air, industrial machine, atomic nucleus, algorithm. If this theory remains valid in the coming years, then the fourth 68 will need to be thought of not simply as a war with an AI component, but as the beginning of an epoch in which the very making of a combat decision becomes a weapon.

This makes the current WAI not just an index for one war, but an early indicator of whether humanity is entering a regime in which the speed of machine-assisted violence begins to outpace the political capacity for control, correction and stopping. If this is confirmed, the fourth 68 will not be just another case in a series. It will be the first full algorithmic threshold of modern warfare.

22. Conclusion, summary and practical direction - what the entire corpus of points 1-21 actually says

The second major conclusion is even more unpleasant. If one traces not only the number but the consequence after the number, an ascending order of pain, death, technological acceleration and systemic rearrangement is visible. The first 68 does not stay in Libya but leads to the Balkans and to the loosening of the European system. The second 68 does not stay between Austria-Hungary and Serbia but destroys empires and prepares an even greater calamity. The third 68 does not stay a European war but rewrites the very limit of destruction through the atomic bomb and the nuclear regime. The fourth 68 already shows the same structure of spillover: from the front in Ukraine to energy, industry, drones, AI, alliance trust, regional tensions and the new nervousness of the global system. Therefore the most sober conclusion is not that "apocalypse is necessarily coming," but that we are already in a corridor in which the probability of a much broader catastrophe, a much deeper fragmentation or a much more painful rearrangement is significantly higher than political language and public psyche are inclined to admit.

The third conclusion is that the weapon must no longer be thought of merely as a means of inflicting damage, but as an accelerator of history. This is one of the most important lines of the article. In 1911 the accelerator enters the air. In 1914 it enters the industrial machine. In 1939 it enters the atomic nucleus. In 2022 it begins to enter the algorithm, the data, the autonomy, the compression of the cycle from perception to strike. This is no longer just a theoretical danger but a publicly visible operational shift, including beyond the Ukrainian theater. It is precisely here that the situation is not rosy, but extremely bleak. Because if the nuclear weapon made possible instantaneous mass destruction, the algorithmic weapon makes possible the mass acceleration of decision, error and escalation. This means that future pain may come not only from one enormous explosion, but from thousands of faster, cheaper, more scalable and harder-to-politically-manage cycles of violence. In other words: we are entering not simply an era of more powerful weapons, but an era of faster war, and faster war almost always means less time for human correction, less time for moral control and a greater chance of catastrophic cascade.

The fourth conclusion is practical and it is the reason "History of the Future" and Gen FUTURE exist. Their task is not to adorn fear and not to produce beautiful apocalyptic phrases. Their task is to name early what does not yet have a name, to formalize what others feel only as anxiety, to make visible the invisible connections between date, weapon, tempo, alliance, collapse and historical spillover, to calculate where public language only gesticulates, and to warn before the system has lost its capacity for self-correction. If this text is useful, its usefulness lies not in "proving" 68, but in showing how to work with a dangerous anomaly without intellectual panic and without cheap contemptuous dismissal. The most concrete direction from here on is the following: 2022 must be monitored not only as a war, but as an open index of rearrangement; the accelerators must be measured, not just the casualties; alliance cohesion, the industrial base, AI penetration into the kill chain, energy vulnerability, cybernetic nervousness and the thresholds of autonomous violence must be tracked; and above all, one must work for the preservation of the ability for future choice, because it is precisely the first thing that dies when war becomes too fast, too automated and too all-encompassing. If the fourth 68 has truly opened a new epoch, then the greatest mistake would be to read it as just another crisis. It must be read as a warning that the world may have already entered the next threshold of historical danger, without yet having found the language to admit it.

23. Awareness, correction and likely next awareness - what humanity actually learns after the 68-thresholds

The most unpleasant but also most accurate conclusion from the historical material so far is that humanity has not corrected any of the previous 68-thresholds in time. It has recognized things after the blow, not before it. After 1911, no early civilizational brake against the aerial dimension of warfare appears. What appears instead is a military reception of the new capability. The world understands that the air is now a combat space, but does not build a sufficiently strong legal and moral barrier to stop the transformation of aerial advantage into later mass bombing logic. Therefore, after the first 68, humanity learns not how to limit the new accelerator, but how to further develop it.

After 1914, the awareness is deeper but still insufficient. World War I compels states to acknowledge the reality of industrialized mass destruction, of chemical warfare, of total mobilization and of the fact that old diplomacy can no longer alone sustain the tempo of modern conflict. The League of Nations appears, collective security as an idea and the Geneva Gas Protocol as an attempt to place at least a partial barrier against chemical horrors. But the correction is incomplete and weak. Humanity recognizes the weapon but does not sufficiently correct the political order that produces the next catastrophe. Therefore, after the second 68, the learned lesson is real but partial: certain means are limited, without sufficiently fixing the architecture of international instability.

After 1939, humanity finally makes its strongest corrective leap. Here we already see not just alarm but an institutional response of a new scale: the UN, Nuremberg, the 1949 Geneva Conventions, and later the nuclear non-proliferation regime. This is the most serious demonstrable historical correction after a 68-threshold. The world no longer merely recognizes the new fatality; it begins to build lasting international mechanisms designed to make the next global catastrophe less likely or at least harder to unfold. But even here, the correction is not complete. The nuclear weapon is not eliminated. It is managed. Therefore even the most successful post-68 corrective in history does not eliminate the accelerator but places it in a regime of fear, deterrence, control and partial legal fencing.

From this comes the most general rule that can be scientifically formulated. Humanity so far has not shown the ability to recognize the new accelerator early enough to stop it before its mass historical deployment. It shows a different ability - to build post-hoc management regimes after the cost has already become unbearable. Put more sharply: humanity does not learn preventively but traumatically. It does not correct before the threshold, but after it. This is precisely what makes the fourth 68 so dangerous, because if the same pattern repeats, then the true awareness will come after much greater pain than today's political language allows.

On this basis, several clear scientific hypotheses can be formulated about what will be recognized after the fourth 68, depending on how the current cycle ends or stabilizes. Part of this awareness is already beginning prematurely, precisely because AI weaponization has become publicly visible even before the end of the conflict.

Hypothesis 1: if the war ends with an armistice, a frozen front or a partial deal without a truly sustainable peace, humanity will recognize above all not the moral but the structural lesson. This lesson will state that cheap autonomy, mass drones, electronic warfare, AI-assisted reconnaissance and industrial capacity have become a permanent layer of modern warfare. In this scenario, the correction will be mainly military and industrial: more production, more air defense, more unmanned systems, more data, more autonomous functions. This would mean that the world will learn the technological lesson but may not learn sufficiently well the political and civilizational lesson. The risk here is accelerated militarization without sufficient ethics and without sufficient legal restriction of algorithmic warfare.

Hypothesis 2: if the conflict ends with Ukrainian survival, relative stabilization of the front and consolidation of the Western and especially European defense architecture, then humanity will recognize five key things. First, that defense is no longer only a matter of armies, but of the industrial base. Second, that alliances cost as much as their real productive and political carrying capacity. Third, that energy security is part of military security. Fourth, that AI in the kill chain cannot remain entirely unregulated. Fifth, that Europe cannot endlessly outsource its own existential risk. In this scenario, the correction would be more mature. It would include not only more weapons, but also a clearer concept of strategic autonomy, AI limitations, industrial resilience and protection of critical infrastructure.

Hypothesis 3: if the conflict ends with a forced deal in favor of the aggressor, lasting erosion of alliance trust, or with the sense that nuclear blackmail and war of attrition work, then the awareness will be much bleaker. The world will learn not the lesson of limitation, but the lesson of fear. This lesson will state that the international order is weaker than it appeared; that guarantees are conditional; that attrition can defeat the law; and that automation, drone mass and the threat of escalation give advantage to actors willing to bear enormous long-term cost. In such a scenario, it is likely to see accelerated proliferation of drone, AI and autonomous combat regimes, an even more nervous nuclear world and a greater temptation for states to seek their own ultimate forms of deterrence. This would be a form of "awareness," but of the worst type - awareness through loss of trust in the very idea of stable order.

Hypothesis 4: if before the stabilization of the conflict a broader regional or inter-bloc escalation occurs, then humanity will probably finally recognize that algorithmic and autonomous warfare are not merely a new means, but a new strategic risk of civilizational order. In this scenario, after the war there would arise pressure for something analogous to the post-1945 correction: harder international regimes for human control over strategic decisions, rules for autonomous systems, protection of critical infrastructure, new mechanisms for de-escalation between nuclear powers and a new type of international control over AI-assisted targeting and combat swarms. This would be the deepest and most mature awareness, but also the most tragic, because it would come only after a much larger catastrophe.

There is also a fifth, more structural hypothesis, which is independent of the specific outcome of the war. It is that after the fourth 68, humanity will almost certainly recognize that tempo has become a central category of warfare. In the twentieth century the primary horror was scale. In the twenty-first century, speed is added to scale. When recognition, targeting, coordination, logistics and strike are accelerated by machines, political control turns out to be chronically belated. If this hypothesis is true, the greatest post-2022 awareness will not be just "AI is dangerous," but something more specific: that civilization must build mechanisms for deceleration, stopping, human verification and correction precisely where war becomes too fast for its own political shell.

Therefore, if summarized as clearly as possible, humanity so far has recognized after the previous 68 thresholds three things: first, that the new accelerator is real; second, that the old rules do not accommodate it; third, that the correction comes too late. After the fourth 68, the most likely great awareness will be that the algorithm, the autonomous system, the industrial tempo, energy resilience and alliance reliability are no longer separate topics but one common architecture of survival. If this lesson is learned early, the world can enter a difficult but correctable regime. If it is learned late, then the fourth 68 will turn out to be not just another date in the history of wars, but the threshold after which humanity finally understood the danger - but again after it had paid with too much death, pain and destruction.

24. The time intervals between the 68-thresholds and the question of the fifth 68

If the four 68 cases are examined not only as separate dates but as a sequence, it is immediately visible that there is no uniform periodicity between them. This is the first and most important scientific result from the temporal analysis. The intervals do not move like a clock, do not form a simple cyclicity and do not allow mechanical forecasting of the type "every X years a new 68 comes." This is precisely why the topic is more interesting but also more dangerous. If there were equal distances, we would be talking about an ordinary periodic structure. Instead we see expanding, strongly unequal intervals, which suggests that 68 does not behave as a calendar rhythm, but rather as a threshold that appears when history reaches a certain level of accumulated instability and a new military accelerator.

The four dates on which the article works are: 29.09.1911, 28.07.1914, 01.09.1939 and 24.02.2022. From them follow three intervals. Between 29.09.1911 and 28.07.1914 there are 1,033 days, i.e. 2 years, 9 months and 29 days. Between 28.07.1914 and 01.09.1939 there are 9,166 days, i.e. 25 years, 1 month and 4 days. Between 01.09.1939 and 24.02.2022 there are 30,127 days, i.e. 82 years, 5 months and 23 days. Even at this rough overview, it is clear that we are talking not about a regular series, but about a sequence with sharply increasing distances. This means that the time between thresholds by itself cannot be read as a simple calendar cycle.

If the intervals are examined as a numerical sequence, the picture becomes even more illustrative. The first increase is from 1,033 to 9,166 days, which is approximately 8.87 times. The second is from 9,166 to 30,127 days, which is approximately 3.29 times. This means there is no stable growth coefficient. If there were a constant multiplier, we could at least conditionally speak of exponential logic. But even that is absent here. In absolute value, the increase is also not stable: first the interval grows by 8,133 days, then by 20,961 days. Therefore neither simple linear nor simple geometric periodicity works. The historical meaning of this is important: the 68-thresholds do not look like regular "clock strikes" but like rarely occurring points of intersection between accumulated systemic risk, a new accelerator and a large-scale historical transition.

This observation also changes the very question "when to expect the fifth 68." Scientifically speaking, a single honest exact date cannot be given. Only a corridor of scenario estimates can be given, each of which is a function of a chosen model. If we use the simplest conservative model - repetition of the last interval, i.e. we assume that the next threshold will appear after another 30,127 days - then the reference point is 20.08.2104. If we use a pure arithmetic mean of the three intervals, we get 13,442 days and a reference point of 14.12.2058. If we use linear regression on the intervals themselves, we get approximately 42,536 days and a reference point of 11.08.2138. If we use a quadratic model of accelerating expansion, we get approximately 63,916 days and a reference point of 22.02.2197. These results are not "four forecasts" but a demonstration of something more important: with so few data points, the different models diverge drastically, which means that any exact date would be a pseudo-scientific pretension, not a reliable conclusion.

Therefore the most honest scientific answer is twofold. First: the fifth 68 cannot be reliably dated solely from the previous three intervals. Second: if a working horizon is nevertheless sought, one must think not of a single date but of a probabilistic window. The widest reasonable window following from simple extrapolations is from the mid-21st century to the end of the 22nd century. This is far too wide to be used as an operational forecast, but it is entirely sufficient to show that the fifth 68, if it appears at all in the same line of historical weight, will almost certainly not be "tomorrow" in the trivial sense of the calendar, but will be the next threshold of systemic rearrangement for which humanity can prepare or fail to prepare.

From the perspective of "History of the Future," there is an even more important correction here. The search for the fifth 68 as a calendar fetish is weaker than the search for the conditions under which it would become possible. This is the key distinction. If 68 is not a clock but a threshold, then the right question is not "on what date will it come" but "under what accumulations will it become possible." And the accumulations are already clear from the previous points: a new war accelerator, higher kill chain compression, greater alliance instability, greater infrastructural vulnerability, greater dependence on data, energy and algorithms, and less time for political correction. In other words, the fifth 68 should not be sought as a date but as the next historical threshold at which a new accelerator and a new systemic fracture will coincide upon a conflict with sufficiently great historical density.

Here comes the most practical conclusion as well. Since the fourth 68 is still underway and is not a closed historical case, any search for a fifth 68 as a near fixed point is conceptually premature. We are still inside the unfolding of the fourth threshold. This means that the first analytical duty is not to guess the next date, but to measure the current transition to its end - to what extent AI, autonomy, drone mass, alliance erosion and the multi-theater crisis will become a permanent regime. If this escalates into a full algorithmic, autonomous and multisectoral war, then the fourth 68 itself may turn out to be so long and so transformative that the question of the fifth 68 temporarily becomes secondary. In the language of "History of the Future," this means the following: we are not waiting for the next date; we are watching whether we are already living in a very early stage of the next epoch.

The most synthesized conclusion is: the intervals between 68-thresholds do not allow clock-like forecasting; they show an expanding and unstable sequence; therefore the fifth 68 cannot be honestly predicted as an exact date, but only as a distant threshold for which conditions must be observed. If a working reference point is still needed, the most conservative structural reference point is around 2104, but this is not a forecast but a reference point in a wide corridor of uncertainty. The true task of "History of the Future" and Gen FUTURE here is not to prophesy a day, but to make visible the early signs of the next threshold while there is still space for correction and preservation of the ability for future choice.

Therefore the fifth 68 should not be sought as a date, but as the next historical threshold at which a new accelerator and a new systemic fracture will coincide.

25. What must be done now, before the fourth 68 has escalated into its darkest form - policies, laws and institutions by "History of the Future"

If the previous three 68 thresholds have taught us anything, it is cruelly simple: humanity almost always builds the necessary institutional architecture after the catastrophe, not before it. After 1911 aerial warfare is recognized but not limited in time. After 1914 come late and partial attempts at collective security and restriction of some weapons. After 1939 humanity finally makes the great institutional leap - the UN, the Geneva Conventions, Nuremberg, the NPT - but again after the mass death. If the fourth 68 is moving toward the darkest scenarios, the most important lesson is clear: if we wait for "sufficient evidence," we will probably wait until the moment when the cost is already unbearable. Therefore, the policy of the mature world must not be reactive, but preemptive.

In the logic of "History of the Future," this means not just "more defense," but the creation of an architecture for preserving the ability for future choice before algorithmic, autonomous and multisectoral warfare has become an irreversible standard. What is needed is not a single measure and not a single new treaty, but an entire package of legal, institutional and industrial instruments that will slow the tempo where machine violence is accelerating, return human accountability where it is eroding, and make societies harder to break where the next war will strike not only the front but energy, logistics, networks, cities, financial systems and political psyche.

The fact that AI acceleration is already publicly visible in more than one theater makes the preemptive institutional response even more urgent.

The first necessary step is a new international legal regime for human control over the use of force. A binding treaty of the type Convention on Human Control over Lethal Force is needed, which explicitly prohibits autonomous systems capable of selecting and striking human targets without meaningful, specific and traceable human control. This regime must be clearer and firmer than the current voluntary declarations. It must contain minimum legal standards: mandatory human verification for lethal decisions; prohibition of autonomous anti-personnel targeting; prohibition of delegating nuclear, chemical, biological and other strategic decisions to algorithmic systems; mandatory capability for abort, override and technical stoppage at every critical stage of the kill chain. If this is not introduced preemptively, the fourth 68 will push the world toward a situation in which "human control" remains a rhetorical, not a legal category.

The second step is a domestic and allied legislative framework for accountability of military AI. A legislative package of the type Algorithmic Warfare Accountability Act is needed, which makes mandatory the following elements: continuous system logs; traceable data provenance; clear distinction between recommendation and execution; independent red-teaming before combat deployment; mandatory reporting of AI incidents; personal command responsibility for deploying a system outside the approved operational regime; technical and legal model cards for every system that participates in targeting, prioritization or autonomous action. Without such a law, the worst of both worlds will occur: machine speed without machine accountability and human formal responsibility without real human control.

The third step is a new international institution, analogous in logic to the IAEA, but for autonomous and algorithmic warfare. The most useful model would be an International Agency for Autonomous and Algorithmic Military Systems - small, technical, inspectional, accountable, but with a strong mandate. It should not deal with abstract ethics, but with verifiable reality: a registry of combat-deployed autonomous systems; auditability standards; independent post-incident investigations; minimum requirements for human override; inspections by invitation or by mandate in high-risk incidents; public reports on classes of systems and prohibited configurations. If the nuclear age imposed the need for an inspection architecture, the algorithmic age requires something similar, but faster, more flexible and more deeply connected to data, software, chips and combat use.

The fourth step is a new convention for the neutrality and protection of critical infrastructure under conditions of algorithmic and hybrid warfare. The Geneva Conventions and existing international humanitarian law remain a foundation, but are obviously insufficient for the new environment in which power transmission networks, submarine cables, satellite channels, water systems, hospitals, cloud centers, financial payments and nuclear safety are simultaneously civilian tissue and strategic targets. A new generation of legal regime is needed, one that defines specially protected categories of infrastructure and introduces firmer rules prohibiting autonomous or AI-assisted attacks on nuclear plants, hospitals, water supply, major energy grids and vital digital communications systems. This is not about a moralistic gesture, but about protecting the very minimum civilizational carrying capacity.

The fifth step is the creation of a joint NATO-EU architecture for resilience, industrial reserve and strategic continuity. The historical lesson of the fourth 68 is that victory or survival no longer depends only on bravery, but on the ability to produce drones, ammunition, spare parts, electronics, sensors, air defense systems and means of repair and recovery at sufficient speed. Therefore, common defense readiness pools, cross-border reserves, coordinated stocks of critical components and clear planning of industrial surge capacity are needed. This must not remain merely a documentary framework. It must become an institutional mechanism with budget, obligations, stress tests and activation thresholds. In the world after 2022, this type of capacity is not an economic extra. It is part of deterrence itself.

The sixth step is building a mechanism for decelerating machine escalation. If the fourth 68 truly introduces the algorithm as a new military accelerator, then civilization needs institutions not only for force, but also for tempo. At least three things are needed here: first, permanent military hotlines and a specialized channel for incidents with autonomous systems between nuclear and major military powers; second, common rules for notification upon an autonomous incident with potential for interstate escalation; third, internationally recognized procedures for emergency pause in case of risk of misattribution, cascade launch or automated erroneous response. In the nuclear age, humanity learned that de-escalation channels are needed. In the algorithmic age, these channels must operate at much shorter time horizons.

The seventh step is democratic and society often underestimates it. If the darkest scenario unfolds, the blow will not be only external. It will enter inside societies through fatigue, division, disinformation, panic, institutional distrust and demands for emergency concentration of power. Therefore a new class of institutions for civic resilience and democratic continuity is needed: protected communication channels during crisis, civic continuity plans, transparency about critical dependencies, training for infrastructural and informational resilience, protection of electoral and governance processes under hybrid pressure. In other words, it is not sufficient to protect borders. The ability of the political community to remain free and coordinated under prolonged pressure must be protected.

The eighth step is the most difficult but also the most important: preemptive red lines for AI in the nuclear and strategic sphere are needed. This means an internationally agreed rule that AI cannot be the last instance in the decision for nuclear use; that early warning systems cannot be linked to autonomous lethal chains; that autonomous swarms cannot be used against nuclear command, communication and control systems; that states are obliged to maintain a clear human chain of responsibility for every action with potential for strategic escalation. If humanity misses precisely this layer, then the next phase of the fourth 68 may prove not merely bloody, but civilizationally uncorrectable.

Within the framework of "History of the Future" and Gen FUTURE, this entire proposal can be reduced to one principle: if the new accelerator is the algorithm, then the answer cannot be just more speed. There must be more law, more inspection, more human accountability, more reserves, more infrastructure protection and more institutions for decelerating the fatal error. After 1911 the world did not do this in time. After 1914 neither. After 1939 it did much, but at a monstrous cost. If the fourth 68 is moving toward its darkest form, then the only mature policy is to build the post-2022 institutions before the post-2022 catastrophe has concluded. Not after. Now. Otherwise we will repeat the historical pattern in its worst variant: we will understand the danger correctly, but too late.

26. What has been overlooked so far - war is no longer only on the map, but in the supporting layer of civilization

The most important thing that has not been stated sharply enough so far is that the fourth 68 does not simply turn Ukraine into a laboratory of new warfare. It turns the very supporting tissue of civilization into a battlefield. In the previous 68 cases, the new accelerator could still be thought of as a weapon positioned relative to the front. In 2022 this is no longer sufficient. The true front runs through the chips, cables, satellites, data centers, logistical corridors, transformers, maritime chokepoints, the extraction of critical minerals, cloud services, image and voice authentication systems and through the very ability of society to distinguish evidence from simulation. If this layer is missed, the entire analysis of war remains too superficial.

This is precisely why conflicts like Ukraine and Iran can no longer be thought of merely as state operations, but as hybrid sovereign configurations between states, models, platforms, satellite and cloud dependencies.

The first gap is that most political and media languages still speak of war as if territory were the primary object of control. This is no longer entirely true. In the world of the fourth 68, control over territory is increasingly a derivative of control over the supporting systems. A state may formally hold borders and institutions, but if it is dependent on narrow semiconductor import chains, on vulnerable submarine cables, on a small number of satellite services, on concentrated mineral supplies and on external cloud infrastructures, its real strategic freedom is much smaller than it appears. This means that in the new war the classical distinction between front and rear begins to collapse. The real rear is now the set of hidden dependencies without which society cannot see, communicate, produce, pay, repair and coordinate.

The second gap is that civilian infrastructure is still thought of as "background" rather than the main target of systemic attrition. This is an intellectual error. Submarine cables, power grids, digital exchange points, satellite services and the chains of communication resilience are no longer secondary support systems. They are the very nervous system of the modern state. If they are disrupted, delayed, contaminated with false data or put in a regime of permanent uncertainty, society can lose more than its ability to respond militarily. It can lose its ability to know what is actually happening. It is precisely here that war becomes more brutal than the old models - not because every strike is more powerful, but because more and more strikes act upon the conditions that make order possible as such.

The third major gap is the crisis of evidence. In the article, AI was already examined as a kill chain accelerator, but this is only half the problem. The other half is that AI undermines the very epistemic environment in which societies and leaders distinguish real from fake. When synthetic video, synthetic voice, manipulated images and mass informational fog become sufficiently good, the crisis is no longer just disinformational. It becomes strategic. It is possible that the next great escalation will arise not from a large army, but from a misattributed video, a fake recording, convincing synthetic testimony or a rapidly spread manipulation that strikes the decision window before there is sufficient time for verification. This means that in the epoch of the fourth 68, truth ceases to be merely a moral question. It becomes critical infrastructure.

The fourth gap is dependence on narrow industrial and mineral chains. Modern war is no longer simply a question of steel, fuel and human reserves. It is a question of chips, optics, rare elements, batteries, magnets, high-frequency components, sensors, memory and production capacities distributed unevenly across the world. This is a radical change. The next great catastrophe may come not only from the decision to shoot, but from the failure to deliver, replace, repair and compute. Therefore the concept of "arms race" must already include a race for chip resilience, mineral security, cable protection, space continuity and authenticity of digital evidence. Whoever does not see this is still thinking in industrial twentieth century terms, while war has already stepped into the infrastructural twenty-first.

The fifth gap is the role of private actors. Previous epochal wars could be thought of primarily as competitions between states. Today that description is insufficient. Private companies hold parts of the communication orbit, the cloud environment, the chip chains, the logistics platforms, the content authentication and the infrastructure support. This means that under the conditions of the fourth 68, the private sector is no longer simply the economic periphery of war. It is one of the carriers of sovereignty. This does not mean that corporations have replaced the state. It means that without them the state often can no longer exercise its own power. From this follows an uncomfortable but necessary conclusion: the future theory of war must think not only of states and armies, but of hybrid sovereign configurations of states, platforms, providers, satellite operators, chip ecosystems and network intermediaries.

The sixth gap is that the public imagination still underestimates the indirect path to mass death. It is easiest to count casualties from missiles, bombs and trenches. It is harder to see what lethal capacity a disrupted electrical system, a shortage of medications, a collapse of cooling, damaged water infrastructure, blocked logistics, a manipulated communications network and substituted evidence have. But it is precisely here that the fourth 68 may prove more terrifying than the previous ones. Because the next great pain may come not only as visible slaughter, but as the disintegrating support of life. This is not a literary metaphor. It is a direct consequence of the fact that civilization is already heavily dependent on mutually connected, fragile and often poorly protected technical systems.

The seventh gap is space. If digital infrastructure and submarine cables are already recognized as critical, space infrastructure still often remains outside the public narrative. This is a serious blind spot. Without satellites, modern communication, navigation, synchronization, reconnaissance, meteorology and part of economic coordination shrink dramatically. This means that the war of the fourth 68 is not only ground, naval and aerial. It is already orbitally dependent. And orbital dependence means a new type of common vulnerability: a small number of assets, very high value, high asymmetry between the cost of destruction and the cost of restoration and an enormous effect of cascading spillover down to life on earth.

From the perspective of "History of the Future," the most brutal but also most useful conclusion is the following: the fourth 68 does not confront us simply with the risk of "another big war." It confronts us with the possibility that war will definitively become a regime of managing civilizational dependencies. In such a regime, victory is measured not only by territory, but by whose cables remain intact, whose chips keep coming, whose satellites are not silent, whose energy does not stop, whose logistics do not freeze, whose evidence can still be believed and whose societies do not internally disintegrate under the pressure of prolonged anxiety. If this is not seen now, then the next catastrophe will be described late and incorrectly. It will be called a "surprising escalation," a "black swan" or a "systemic shock," when in fact it will have been simply the consequence of civilization allowing its supporting layer to become a battlefield without building sufficient legal, industrial and epistemic protection.

Therefore what has been overlooked so far can be compressed into one sentence: in the epoch of the fourth 68, war is no longer simply organized violence upon people and territories, but a struggle for control over the supporting layer of civilization. And whoever wants to be useful today must not only think of tanks, missiles and fronts. They must think of cables, orbit, chips, minerals, logs, authenticity, reserves and tempo. It is precisely there, in this seemingly "technical" layer, that the next great historical vulnerability lies hidden.

27. Final synthesis - what 68 actually revealed upon deep reading by "History of the Future"

In the beginning stood something that could easily have been dismissed as an internet meme, a number game or conspiratorial noise. Three dates, one number, a quick pattern, easy sensation. At a superficial reading, the topic looked like yet another symmetry upon which the collective imagination projects meaning. But upon a systematic, disciplined and scientifically constrained approach, it turned out that beneath this seemingly light entry there was something more serious. Not a proven cosmic code, not a finished law of history, but a real anomaly of high analytical value. The unexpected was not that a number exists. The unexpected was that the number stands upon four thresholds of war with unusually high historical density, and that each of these thresholds coincides not only with a large conflict, but with a displacement of the very nature of war toward a new accelerator, a new scale and a new regime of subsequent rearrangement.

This is the most important thing that became visible from point 1 to point 26. 68 did not turn out to be interesting because it is a rare number. It turned out to be interesting because in the observed corpus it four times marks start dates after which the world is no longer the same. 1911 is not simply colonial aggression, but the entry of the aerial perspective and the verticalization of warfare. 1914 is not simply the start of a great European conflict, but the collapse of the old architecture and the birth of industrial mass destruction. 1939 is not simply another world war, but the threshold within which the nuclear limit is born. 2022 is not simply the continuation of a geopolitical dispute, but an open epochal point at which the algorithm, autonomy, data, kill chain tempo, critical infrastructure and alliance instability begin to act together as a new type of systemic risk. The new public data already make this claim less hypothetical: the algorithmic accelerator is visible not only as a possibility, but as an operational reality. Thus the original "pattern" brought us to something far more important than the number itself - to a theory of the historical thresholds of acceleration.

In this sense, the greatest understanding that was achieved is the following: the true problem was never 68 in itself. The true problem is what 68 so far coincides in marking - moments in which a new accelerator enters war and makes it faster, deeper, harder to stop and more dangerous for the very civilizational carrying capacity. From there emerged the heaviest conclusions of the text: that humanity so far has not corrected any of these thresholds in time; that it has learned post-hoc, not preventively; that the weapon evolves from object to regime; that in the fourth 68 it is no longer only armies fighting, but entire supporting layers of civilization; that truth, logistics, chips, cables, satellites, minerals, energy and human control over the decision have become part of the battlefield. This knowledge is not pleasant. But it is useful precisely because it is unpleasant. It shifts the conversation from sensation to architecture.

Even more useful was something else. Instead of dissolving the topic in mysticism, it forced the analysis to become more disciplined. Calendar normalization was required. Restriction to a precisely dated corpus was required. Distinction between phases of war and between symbolic and operational beginnings was required. The creation of an index, a theory, working hypotheses, scenarios, institutional proposals and corridors of risk was required. This is also the essential usefulness of "History of the Future" and Gen FUTURE. They are not here to adorn anxiety, but to translate it into a thinkable form. To take something that begins as a meme and bring it to the point where it becomes an instrument for seeing, for warning, for formalization and for early preparation. When the public culture either mocks or hysterizes, the useful labor is a third option: to see, to name, to arrange, to calculate, to outline boundaries and to show what must be done now.

This is precisely where the text reaches its most important warning. If the fourth 68 is truly the threshold of algorithmic and autonomous warfare, then the danger facing civilization is not just "another big war." The danger is the transition to a world in which the tempo of violence, the automation of decision, the vulnerability of infrastructure and the crisis of evidence begin to outpace the human capacity for political correction. In such a world, catastrophe is no longer only a question of one army or one bomb. It becomes a question of an erroneous signal, a damaged cable, an accelerated swarm, disrupted logistics, false evidence, an exhausted alliance and a political system that loses time faster than it gains it. This is the essential darkness of the fourth 68. Not that it is "fatal" in a poetic way, but that it may turn out to be the first threshold at which war enters directly into the nervous system of civilization.

From this follows the final practical usefulness of the entire analysis. If all of this is true even by half, then there is no point in waiting for a fifth 68 as a superstitious date. The point is elsewhere - to measure the current acceleration, to limit autonomous lethal systems, to preserve human control over force, to protect critical infrastructure, to strengthen industrial and energy resilience, to stabilize alliances, to create regimes for decelerating machine escalation and above all to protect truth as infrastructure. If this text has reached this far, then its usefulness is not in convincing everyone that the number 68 is mystical, but in showing that beneath the surface of a meme there can stand a genuine historical question: at which threshold is civilization and is there still time to correct course, before the acceleration has become irreversible.

28. Protocol for observing the fourth 68 - how to recognize its transition into the most dangerous phase

After points 1 through 27, it is no longer sufficient to say that the fourth 68 is "dangerous." This is true, but analytically insufficient. According to "History of the Future," the next necessary step is to create an observation protocol - not to predict with false precision the day of the next catastrophe, but to make visible when the current threshold transitions from a severe war to a new civilizational regime of risk. In other words: if the fourth 68 is still open, we need not more metaphors, but an observational framework by which to distinguish whether the system is still correctable, or is already moving toward a self-sustaining cascade.

This protocol must work not with a single indicator, but with simultaneous observation of multiple layers, because it is precisely the multiple layering that distinguishes the fourth 68 from the previous thresholds. If 1911 can be grasped through the air, 1914 through the industrial machine and 1939 through the nuclear limit, 2022 cannot be read through a single carrier. Its true danger lies in the connectivity: AI, drones, alliances, infrastructure, energy, evidence, the Arctic (as a zone of strategic sensitivity), the Middle East, industrial capacity and political fatigue. Therefore the observation protocol must be built not upon a single alarm, but upon an architecture of alarms. In this analysis, the Arctic is not examined as an already unfolded major war, but as a growing field of strategic sensitivity, alliance tension and infrastructural risk, which expands the geography of the fourth 68 beyond its initial theater.

The first observed layer is the diffusion of the algorithmic kill chain. The key question here is not whether AI is used on the battlefield - this is already a fact. The question is whether AI remains an auxiliary layer, or transitions into a normalized operational standard for target generation, prioritization, battle planning and post-strike assessment in more than one theater. When Ukrainian combat data are opened for training of allied AI models, and in the campaign against Iran the role of Maven + Claude in the targeting workflow is already publicly described, this is already a signal that the accelerator is not remaining local. Here the protocol must track three thresholds: first, cross-theater diffusion; second, institutionalization in doctrine; third, displacement of human judgment from the center to the periphery of the combat decision.

The second layer is the normalization of autonomous mass. The war in Ukraine has already shown that cheap drones, autonomous interceptors and semi-autonomous systems are not a temporary improvisational layer, but the beginning of a new economy of violence. When the largest European military states begin urgently working on low-cost autonomous means for air defense and counter-drone protection, this means that the lesson from Ukraine is transitioning from observation to production. Here one must track whether autonomous systems remain defensive and limited, or gradually transition toward swarming, semi-autonomous and mass-reproducible attacking logic. This transition would be one of the most dangerous signals that the fourth 68 is beginning to self-accelerate technologically.

The third layer is the theatrical coupling of otherwise different conflicts. The most important question here is not whether Ukraine, Iran, the Arctic and the resource disputes are "one war" in a juridical sense. They are not. The question is whether they begin to act as one system in a strategic sense. If resources, attention, allied will, military capacity and industrial decisions begin to be redistributed not along separate lines but in a regime of general pressure, this means that the world is already in a polycentric crisis regime. Signals for this include, for example, the tension around Hormuz, the allied divergences over the Middle East, the parallel assurances that Ukraine "will not be abandoned," and the growing significance of the Arctic as a zone of alliance sensitivity, infrastructural risk and strategic presence. When the system must sustain too many fronts simultaneously, it enters a zone in which even a limited error can have a disproportionate effect.

The fourth layer is the cohesion of alliances. According to "History of the Future," this is a critical indicator because the fourth 68 can become most dangerous not necessarily when the adversaries are strongest, but when alliances become most unclear. If Europe arms itself but simultaneously doubts the long-term predictability of the United States; if NATO assures that it will hold Ukraine, but in other theaters political refusals and distancing begin; if the general order of guarantees turns into a negotiation zone rather than a stable framework, then the risk increases even without a direct military breakthrough. The protocol here must track not declarations, but behavioral indicators: real deployment choices, refusals to participate, delays in coordination, different national lines regarding the same crises and the degree to which allies begin to act as separate strategic units.

The fifth layer is infrastructural nervousness. Submarine cables, satellites, energy grids, cloud environments, logistical nodes and financial payments are no longer "background." They are the field of carrying capacity. If the fourth 68 develops toward its most dangerous form, we will recognize it not only by more strikes, but by an increase in pressure on the systems that keep society coordinated. Here the observation must be directed toward three types of signals: first, frequency and severity of infrastructural incidents; second, degree of dependence on a small number of external providers; third, delay or failure in building reserves, redundancy and rapid recovery regimes. If these indicators worsen, it means the fourth 68 is leaving the theater of war and entering the supporting layer of civilization.

The sixth layer is the crisis of evidence. This is perhaps the most underestimated indicator, and in the algorithmic epoch it may turn out to be one of the most decisive. When synthetic content becomes difficult to distinguish from real, when AI enters simultaneously into analysis, targeting and information operations, and when political decisions must be taken ever faster, then the very ability to establish what has happened becomes a strategic vulnerability. The protocol here must track not just deepfakes as a cultural problem, but the systemic erosion of reliable military and political evidence. The moment when public trust in the verifiable event begins to crumble is the moment when the fourth 68 becomes dangerous not only as a war, but as an epistemic collapse.

The seventh layer is the legal lag. If the technical and operational reality moves faster than the law, then the very system of limitation begins to fall fatally behind. Therefore one must track not only whether there are debates about autonomous weapons, but whether a real binding instrument appears that prohibits unpredictable AWS, preserves meaningful human control and introduces clear restrictions on systems used against humans. If even at the current tempo of AI weaponization the international framework remains at the level of declarations, meetings and ethical wishes, this will be a strong indicator that the fourth 68 is developing according to the classic human pattern: technologies go forward, and norms catch up after the mass cost.

The eighth layer is internal political resilience. No major war develops only externally. It penetrates into budgets, into fatigue, into media, into polarization, into fear and into trust. Here one must track whether the prolonged crisis produces in democratic societies higher resilience, or higher readiness for emergency, authoritarian temptation and strategic fatigue. The fourth 68 will become truly dangerous if it simultaneously accelerates technologically, expands theatrically and corrodes internally the political capacities for reasonable, coordinated, long-term response.

From here, a clear phase map can also be constructed. Phase A is the phase of diffusion: the new accelerator leaves the initial theater and begins to appear elsewhere. Phase B is the phase of coupling: different crises begin to behave as a mutually reinforcing system. Phase C would be the phase of lock-in: when the algorithmic kill chain, autonomous mass, infrastructural nervousness, alliance erosion and the legal lag already form a self-sustaining historical configuration. It is precisely this phase that must be prevented, because after it the correction becomes much more expensive, and sometimes belated.

Based on the available data, the world is already at least in late Phase A and probably in early or mid Phase B.

Therefore the scientific meaning of this point is not to "guess the future," but to give an instrument for early reading of the present. If the fourth 68 is ever described retrospectively as the beginning of a new epoch of warfare, this will not be because the number was magical. It will be because even while it was unfolding, it was showing multiple measurable signs that the world was transitioning into a faster, more connected, more algorithmized and more nervous form of conflict. The true usefulness of "History of the Future" and Gen FUTURE here is not to wait for tomorrow's historians to say what was visible. The usefulness is to make the visible visible in time.

29. Criterion for correctability of the fourth 68

If point 28 answers the question of how to observe the fourth 68 in real time, then point 29 must answer the harder question: how will we know whether it is still correctable. This is the decisive threshold in "History of the Future." It is not sufficient to know that the risk is growing. It is necessary to know whether the system still possesses the ability to stop, to slow, to limit, to return human control and to preserve the ability for future choice, or has already entered a regime in which the acceleration of conflict outpaces the political, legal and institutional capacity for intervention.

Here "correctability" does not mean optimism, nor a moral desire for peace. It means something much more specific: the ability of the international system and of the key political communities to limit the new accelerator before it has become irreversibly normalized. In the case of the fourth 68, this means the ability to limit the algorithmic, autonomous and infrastructure-dependent regime of warfare before it becomes a self-sustaining environment in which human decision, law and allied coordination act only post-hoc. Therefore correctability is a threshold capacity for timely intervention, not an abstract virtue.

In the shortest terms, the fourth 68 remains correctable if five supporting systems can still be limited politically and institutionally: human control over the use of force, alliance cohesion, the resilience of critical infrastructure, the legal capacity for prohibition and limitation of new weapons and the public ability to distinguish fact from simulation. If these five supporting systems stabilize, the acceleration can be slowed. If they break down simultaneously, the system enters a zone of hard-to-reverse escalation.

The first criterion for correctability is human control over the lethal decision. This is the most fundamental test. If the international environment manages to impose meaningful and traceable human control over autonomous and AI-assisted lethal systems, then the fourth 68 remains at least partially correctable. But if a practice normalizes in which the model, the sensor network and the automated system begin to determine targets, time windows and priorities without a real, specific and accountable human layer of judgment, then correctability begins to shrink dramatically. This is precisely why the international calls for a binding instrument on autonomous weapons are not peripheral ethics, but a central test of whether civilization can still limit its own new accelerator.

The second criterion is the speed of legal reaction relative to the speed of technological diffusion. If AI, autonomy and drone mass are institutionalized faster than treaties, prohibitions, restrictions and accountability regimes appear, then law is no longer regulating the historical process, but only documenting its delay. The fourth 68 remains correctable only if the legal framework ceases to be chronically reactive. This means not another declaration, but a clear international mechanism with two elements: first, prohibition of unpredictable or inherently uncontrollable autonomous systems; second, restrictions on all other high-risk military AI configurations such that human responsibility is not formal but operational.

The third criterion is the resilience of critical infrastructure. In classical wars, one could distinguish the front from the base of society. In the fourth 68, this distinction erodes. Cables, satellites, energy grids, cloud platforms, financial and logistical systems are no longer just background, but the supporting tissue of the very military and political capability. Therefore the fourth 68 remains correctable only if there is real, not symbolic progress in redundancy, back-up planning, civil preparedness, resilience spending and transnational protection of critical nodes. If societies remain heavily dependent on narrow and vulnerable chains without reliable reserves, then even a limited conflict can produce a disproportionate systemic collapse.

The fourth criterion is the cohesion of the alliance architecture. This is simultaneously a military and political indicator. Even a powerful technological accelerator can be limited if the supporting alliances are coordinated, predictable and industrially backed. But if within NATO, the EU and the broader Western formats trust begins to disintegrate regarding who protects whom, at what risk, with what time horizon and at what cost, then the fourth 68 begins to shift from a correctable crisis to a self-expanding historical environment. What is decisive here is not the quantity of declarations, but whether there is resilience of decisions under pressure from simultaneous theaters, energy shock, the Arctic, the Middle East and internal political fatigue.

The fifth criterion is epistemic resilience. This is the most underestimated layer, but in the algorithmic epoch it may turn out to be decisive. If societies and states lose the ability to verify events, to distinguish actually occurred strikes from synthetic manipulation, to sustain trust in investigation and verification, then correctability falls even in the presence of formally good institutions. The reason is simple: the system cannot correct itself if it cannot reliably describe what has happened. Therefore the fourth 68 remains correctable only if truth remains institutionally defensible, not merely culturally desired.

From here a threshold scheme with three regimes can be constructed.

Regime I - correctable acceleration. This is a state in which the new accelerator is already visible and dangerous, but has not yet become fully normalized and self-reproducing. The characteristic signs here are the following: human control is still real in critical lethal decisions; international law lags but is already moving toward a binding instrument; alliances still function as coordinated systems; infrastructural vulnerability is recognized and real strengthening is beginning; and the crisis of evidence has not yet destroyed the possibility of a common factual framework. In this regime the fourth 68 is dangerous, but not yet lost.

Regime II - hard-to-correct acceleration. This is a state in which at least three of the supporting systems deteriorate simultaneously. For example, AI is already durably in the targeting workflow in more than one theater; autonomous and semi-autonomous systems are proliferating faster than normative control; alliances become visibly more conditional; critical infrastructure is under permanent pressure; and law fails to become a binding limiter. In this regime the system can still be slowed, but the cost of correction becomes high, and the window for political action begins to close.

Regime III - threshold of uncorrectability. This does not mean complete physical collapse or necessarily a world war. It means something more structural: the accelerator has already become an environment. In such a regime, autonomy, AI-assisted violence, infrastructure warfare, the crisis of evidence and the erosion of alliance guarantees begin to reinforce each other. Law lags durably. Political systems act urgently, not strategically. Societies habituate to permanent mobilization and high anxiety. War is no longer simply an event, but a regime of civilizational organization. If the fourth 68 enters here, it will be much harder to reverse than the previous thresholds, because the new accelerator will not be a single weapon, but a general ecosystem of speed, data, automation and dependence.

Therefore the most practical question is: where on this scheme is the world at present. The most honest scientific answer is: probably in the late part of Regime I and the beginning of Regime II. This means that the fourth 68 is no longer merely an observed anomaly. It is an active historical process with sufficiently many visible signs of diffusion, coupling and acceleration to require not academic distance, but preemptive institutional work. But at the same time there is still evidence that the system has not entirely crossed the threshold of uncorrectability: there are real international efforts for binding restriction of autonomous weapons, there is visible strengthening of European and NATO resilience, there is accelerated recognition of the infrastructural risk and there is at least partial awareness that AI and the nuclear factor cannot be allowed to approach each other without limiters.

According to "History of the Future," from here follows the strictest conclusion. The correctability of the fourth 68 is not a metaphysical question. It is a historical function of five things: human control, legal tempo, alliance cohesion, infrastructural resilience and epistemic stability. If these five lines are strengthened, then even a very dangerous threshold can be limited. If they weaken together, then the very ability for future choice begins to shrink. And then the greatest mistake will not be that we did not see the risk, but that we saw it yet continued to act as if we still had infinite time for correction.

30. The right to deceleration - a civilizational principle against algorithmic warfare

If the fourth 68 is the threshold at which the accelerator of war enters the algorithm, then the next fundamental question is not only what weapon appears, but what happens to time. In previous epochs, war became more powerful. In the fourth 68, it also becomes faster in the very structure of the decision. Observation, sorting, recognition, prioritization, resource allocation, targeting and strike assessment begin to compress into a cycle that increasingly leaves less room for human judgment, for legal verification and for political de-escalation. It is precisely here that the need arises for a radical but scientifically defensible principle: the right to deceleration.

The right to deceleration is not a sentimental plea for war to be "more humane." It is a civilizational protective mechanism against a specific type of risk: the risk that the tempo of machine-assisted violence will outpace the tempo of human correction. If this happens, law, politics, proportionality, precautionary measures and even factual verification remain formally available but practically belated. Therefore the right to deceleration should be understood as a legal and institutional requirement that in certain classes of high-risk situations, mandatory time must be created for human verification, legal assessment, allied coordination and epistemic checking.

This is radical because it shifts the center of regulation. Until now, modern limitation regimes have focused mainly on the type of weapon, the carrier of destruction and the permissible targets. In the epoch of the fourth 68 this remains necessary, but is no longer sufficient. The new problem is tempo as an independent source of strategic danger. A system may formally include a human, but if it leaves them seconds or minutes for judgment over a huge volume of algorithmically prioritized targets, the control becomes nominal. Therefore in the new warfare not only "what" and "with what" must be regulated, but also "how fast" it is at all permissible to reach lethal action.

For this reason, the right to deceleration must be formulated as a general civilizational principle: when the combination of autonomy, AI-assisted targeting, infrastructural vulnerability, possible escalation between alliances and unclear evidence exceeds a certain threshold, the system is obliged to activate a deceleration regime. This regime does not mean passivity. It means controlled extension of the time for judgment where the error would be disproportionate, multisectoral or irreversible.

The first application of this principle is in lethal autonomous and semi-autonomous systems. If a system has the capacity to independently recognize, select and strike human targets or critical infrastructure, then the right to deceleration requires a mandatory human review window before the strike. This review window must not be symbolic. It must be sufficient to allow real legal and factual assessment, not merely a formal pressing of a button. Otherwise "human in the loop" becomes rhetoric, not a limiter. Therefore the right to deceleration is a natural extension of the idea of meaningful human control, but develops it further: it is not sufficient for the human to be present; they must also have time in which their presence has content.

The second application is in strategic and nuclear systems. Here the right to deceleration must become an absolute principle. Any integration of AI in command, control and communications systems that shortens the reaction time or increases the pressure on decision-makers in a nuclear environment increases the risk of catastrophic error. Therefore the right to deceleration in this layer must mean: prohibition of AI being the last instance in decision-making for nuclear use; prohibition of automated coupling between warning systems and lethal action; and mandatory multi-person and multi-channel human verification procedures in every scenario with potential for strategic escalation. Here deceleration is not weakness. It is the last barrier against systemic error that cannot be repaired.

The third application is in cyber and infrastructural incidents. The fourth 68 has already shown that war moves through cables, satellites, clouds, energy systems and digital authentication. This means that a single major disruption, a manipulated signal or a massive infrastructural incident can produce pressure for a very fast retaliatory response, often with incomplete clarity about who the perpetrator is and what the actual target is. Here the right to deceleration must mean mandatory pause protocols after certain classes of incidents: time buffers for attribution review, allied consultation, technical verification and legal assessment before transitioning to a highly escalatory response. Without such a buffer, civilization risks reacting with the speed of its network nervousness, not with the speed of truth.

The fourth application is in target generation systems. When AI systems begin to process enormous datasets and propose long lists of potential targets, a quantitative and moral asymmetry appears. The machine can produce hundreds of proposals quickly, but the human cannot give them hundreds of separate full moral-legal assessments at the same pace. It is precisely here that the risk of "human judgement" being displaced by "human confirmation" is born. Therefore the right to deceleration must introduce a limit on the pace of machine-generated target mass: the larger the AI-produced target volume, the higher the threshold of review, sampling audit, independent verification and traceable accountability must be.

The fifth application is allied and international. If the world is entering a polycentric environment in which Ukraine, Iran, the Arctic, critical infrastructure and AI are strategically coupled, then the right to deceleration must also have a coalition form. This means mechanisms for emergency consult-before-escalate between allies, rules for notification upon an autonomous incident, minimum time windows for intra-alliance assessment of major AI-assisted strikes on critical objects and protocols for shared verification of events with strategic weight. Without such a layer, every coalition becomes a risk system in which one accelerated national cycle can drag all the others.

Here the most important scientific conclusion also appears. The right to deceleration does not contradict effectiveness. It distinguishes two types of effectiveness. The first is narrowly military - to strike faster. The second is civilizational - to avoid escalation that destroys the supporting conditions of order. In the epoch of the fourth 68, these two types of effectiveness no longer coincide automatically. It is possible for a system to become more effective in the military sense and more destructive in the civilizational sense. This is precisely why the right to deceleration must be understood as a corrective to a dangerous intellectual distortion: that the faster decision is always the better decision. Under conditions of algorithmic warfare this is often untrue.

According to "History of the Future," the right to deceleration must also be thought of as a principle of preservation of the ability for future choice. If the system reaches the strike so fast that human judgment, law and evidence are left behind, then the future choice shrinks. It shrinks not because it formally lacks, but because it practically can no longer be exercised in time. Therefore deceleration at the critical points is not an anti-modern action. It is a way for civilization to protect its own correctability against a weapon whose most dangerous characteristic is not only the destruction, but the speed of its organization.

31. If the correction fails - minimum architecture for civilizational survival

After points 28, 29 and 30, the hardest question in the entire text can now be formulated. If despite the observation, despite the institutional response, despite the right to deceleration and despite the attempt at correction, the fourth 68 nonetheless develops along its darkest lines, what must remain alive so that civilization does not collapse beyond the threshold of recoverability. This is not an apocalyptic fantasy but a pragmatic question. The historical lesson from great fractures is that societies often do not perish at once. They perish when they lose the supporting systems that make possible coordination, care, evidence, governance and recovery. Therefore, if the correction fails, the task is no longer to "win" in the classical sense, but to preserve the minimum architecture in which life, order and future choice do not fully disappear.

According to "History of the Future," this means not simply thinking about survival, but about carrying capacity under strike. The minimum architecture for civilizational survival is not a romantic vision of post-crisis adaptation. It is an ordered list of those functions that must remain operational even when many other systems are already affected, so that war or systemic crisis does not become a definitive contraction of civilizational space. This architecture does not begin with tanks, but with the supporting layers without which society ceases to be a society.

The first supporting layer is truth as operational infrastructure. Under the conditions of the fourth 68, truth is no longer just an epistemic or moral ideal. It is a practical condition for the continuity of governance, for the limitation of panic, for the avoidance of misattribution and for future accountability. If during a major crisis there are no protected channels for verification, reserve archives, traceable logs, independent authentication nodes and mechanisms for separating document from simulation, society loses the ability to understand what is happening to it. And when it cannot understand what is happening to it, it can neither defend itself, nor bear responsibility, nor recover justly afterwards. Therefore the first minimum task of civilizational survival is for some verifiable line of facts to remain alive.

The second layer is energy. It is not necessary for the entire energy system to remain in normal mode for society to survive. But it is necessary for island energy regimes, backup power, local buffers and a capacity for prioritizing vital nodes to exist. Hospitals, water supply, communication nodes, basic state coordination, emergency logistics and cooling of sensitive infrastructure cannot depend entirely on a perfectly functioning central grid. This means that minimum carrying capacity requires not just "more energy," but energy segmentation, reservation and the ability to maintain critical functions in a regime of prolonged partial damage.

The third layer is communication. If submarine cables, terrestrial networks, cloud platforms, radio relay links and satellite dependencies collapse without backup regimes, society loses not just the convenience of connectivity, but the ability to coordinate its survival. Therefore the minimum architecture requires communication multilayering: central networks, backup state channels, local emergency networks, offline fallback regimes and sufficient autonomy of key communication nodes. Communication in such a context is not technological comfort. It is the foundation of every subsequent response, including health, governance, delivery of water, food and security.

The fourth layer is health. If hospitals, emergency services, trauma care, medication supplies, basic cooling, laboratory logistics and the protection of medical units collapse, mass death comes not only from the strike, but from the inability to treat after it. Therefore the minimum architecture must include functioning protected medical cores, mobile and backup capacities, autonomous stocks of vital medications, operational and triage regimes, and physical and legal protection of health structures. It is not sufficient to protect "the health system" as an abstraction. It is necessary to know which exact functions must survive the first wave of collapse so that a secondary wave of dying does not follow.

The fifth layer is water, food and the basic logistics of life. Here minimalism is decisive. Society cannot preserve its entire economic rhythm during a deep crisis, but it can and must preserve minimum vital continuity. This means local and regional reserves, short chains, functioning nodes for purification and distribution of water, critical food stocks, protected cold-chain capabilities and basic transport organization. Historically, it is precisely the collapse of these systems that often turns a great crisis into prolonged civilizational exhaustion.

The sixth layer is governance. If the state, municipalities and key institutions lose the ability to act in a coordinated fashion, then the remaining reserves very quickly lose their meaning as well. The minimum architecture requires continuity-of-governance protocols: backup administrative channels, physical and digital redundancy of governance functions, human override during digital failure, clear delegation of roles upon destruction of communication and a sufficiently simple, resilient structure for maintaining order. The principle here is clear: the complex state can turn out to be a fragile state if it lacks a simplified regime for continuation of governance under strike.

The seventh layer is law and accountability. This is easy to underestimate in a moment of existential anxiety, but that is precisely when it becomes most important. If the system does not leave a trace of who acted, by whose order, with what means, upon what target and with what consequence, post-crisis recovery almost inevitably degrades into arbitrariness, revenge, falsification and historical fog. Therefore the minimum architecture for survival must include protected archives, geographically and technologically distributed storage of evidence, independent registries of key events and a minimum of procedures for recording the cause, not only the result. If no trace remains, no possibility for justice remains either.

The eighth layer is future choice as the last core of civilizational carrying capacity. This is the deepest but also the most practical layer. A system can temporarily survive physically and yet lose its future, if its entire institutional and moral geometry locks into permanent emergency, permanent mobilization and permanent automation of violence. Therefore the minimum architecture for survival must protect not only life, but also the space for the return of politics, law, correction and de-escalation. If after the crisis there is nowhere to begin the return path, then what survives is not civilization, but only a prolonged regime of contracted existence.

From here follows a strong but sober conclusion. If the correction fails, the most important question will not be "how to return the old world." More likely, it can no longer be returned to its previous form. The true question will be: what minimally must remain operational so that the possibility of recovering order, freedom, truth and future choice is not interrupted. This is the meaning of the minimum architecture for civilizational survival. It is not a plan for victory. It is a plan for society not to fall below the threshold beneath which it can no longer return to human and political form.

According to "History of the Future," this is also the most brutal usefulness of this point. The fourth 68 can still be corrected. But if it is not, civilization has no right to enter such an environment without having first decided what is the irreducible core that must remain alive. Truth, energy, communication, health, water and food, governance, accountability and future choice - these are not eight convenient categories. These are eight tests of whether society is still a civilization, or already only territory after a strike.

32. The black box of war - mandatory combat logs, evidentiary chain and the right to verifiability

If the fourth 68 is the threshold at which war enters the algorithm, then one of the most dangerous results is not only the acceleration of the strike, but the disappearance of the verifiable trace of the decision. It is precisely here that the black box of war opens. In classical wars, one could, at least in theory, reconstruct who made the decision, what order was given, what was seen, what was known and what followed. In the epoch of AI-assisted targeting, data fusion, model-based prioritization and semi-autonomous systems, this reconstruction begins to break down. Not because there is no data, but because the data is too much, the chains are too complex, the tempo is too high and the architecture of the decision is increasingly less transparent to those who formally bear responsibility.

This creates a new class of civilizational risk. If after a strike, incident or escalation it cannot be reliably reconstructed what data entered the system, what the model proposed, what the operator saw, what the commander authorized and how the check for legality was performed, then war begins to lose not only human control, but also historical, legal and moral traceability. In such a world, it is no longer sufficient to say "the human was in the system." It must be possible to prove in what way they were in the system. Otherwise human control becomes a ritual phrase, not a real limiter.

Therefore in the epoch of the fourth 68, a new principle must be formulated: the right to verifiability. This right is not an individual right in the narrow sense, but a civilizational requirement. It means that in every high-risk AI-assisted or autonomously supported lethal process, a sufficiently reliable, protected, traceable and independent trail must remain that allows ex post reconstruction of the decision. If such a trail is absent, not only is justice after the event impeded. Also impeded is the very ability of the system to learn, to limit repeatable errors, to restore trust and to prevent subsequent escalations.

In this sense, the combat black box must be understood not as a metaphor, but as a new minimum standard of legitimacy. Just as aviation accepts that in a crash there must be a reliable record of the critical parameters of the flight, so algorithmic warfare must accept that in a critical AI-assisted decision there must be a reliable record of the critical parameters of the kill chain. Not because this automatically makes war legitimate, but because without such a record, accountability itself becomes nonfunctional.

Practically this means that every high-risk AI-assisted combat system must generate at least five types of trail.

First, a data provenance trail. It must be clear what types of data entered the system - sensor, intelligence, open source, operational, historical, synthetically augmented - when they were collected, from whom they came, with what reliability assessment and whether they were combined with other sources. Without data provenance, there is no way to distinguish whether the decision was based on reliable observation, on a partial picture or on a contaminated and manipulated environment.

Second, a model trail. It must be reconstructible which model was used, which version, what fine-tuning it underwent, what guardrails were active, what confidence thresholds were set, what known failure modes were present and what red-team or pre-deployment tests were done. Without model provenance, there is no way to assess whether the system acted within its known limitations, or was deployed in an environment for which it was not validated at all.

Third, a decision trail. This is the core of the combat black box. Not only the final result must be preserved, but the decision trail itself: what the system recommended, what alternatives it ranked, which targets it excluded, how it prioritized, what the human operator saw, what they changed, what they overruled, what they approved and at which moment. Only this way can machine recommendation be distinguished from human authorization and can it be seen whether the human exercised real judgment or merely confirmed with high speed an already prepared machine flow.

Fourth, a legality trail. In the epoch of AI-assisted targeting, a technical trail is not sufficient. A legal trail is also needed. This means a record must remain of how distinction between military target and civilian object was assessed, how proportionality was considered, what precautions in attack were taken into account, what civilian harm estimate existed, what uncertainty flags were activated and what grounds were used for the conclusion that the strike was permissible. Without this layer, law is reduced to a post-hoc commentary on a technical process, rather than being part of the process itself.

Fifth, a post-strike trail. This includes weapon release data, actual strike timing, battle damage assessment, the difference between the predicted and actual effect, post-strike signals for civilian harm, subsequent corrections and any abort or suspend signals for subsequent actions. If the system does not preserve a post-strike trail, it cannot close its own cycle of accountability. Only the moment of the strike remains, without a true ability for learning and correction.

But this is not sufficient. The combat black box must not merely exist. It must be protected. This means tamper-evident logging, cryptographic signing, trusted timestamps, geographically and institutionally distributed storage, clear separation between operational access and investigative access, and procedures for emergency preservation in case of incident. Otherwise the system will generate logs that the first powerful institution will simply be able to erase, substitute or isolate from review. Then the black box becomes a simulation of accountability, not its real architecture.

From here follows the radical normative principle of this point: for certain classes of high-risk AI-assisted operations, a rule of the type "no combat black box, no legitimate use" must apply. This does not mean that every combat action can wait for an ideal technological environment. It means that systems that cannot provide minimum verifiability must not be admitted to autonomous or algorithmically assisted lethal action against humans and critical infrastructure. In other words, the right to verifiability must become a gate condition, not a post-hoc wish.

This idea also has an institutional consequence. If the combat black box becomes a standard, a new architecture of access will be needed. Not all data can be public. But there must be clearly defined regimes under which competent military inspectors, parliamentary bodies, international missions, courts or special investigative mechanisms can under certain conditions gain verifiable access to the decision trail. Without such access, the log will remain merely the internal memory of the state or company, not an instrument for real accountability.

Here the connection with the right to deceleration is also visible. Deceleration and verifiability are two sides of the same civilizational mechanism. Deceleration protects the time for human judgment before the strike. Verifiability protects the trail for human and institutional accountability after the strike. If one is missing, the other weakens. If there is no time, the human does not really control. If there is no trail, the human cannot really be verified. In this sense, the fourth 68 will be correctable only if both the right to deceleration and the right to verifiability are built simultaneously.

According to "History of the Future," the deepest meaning of this point is the following: in the epoch of algorithmic warfare, civilization cannot rely only on good intentions, promises of ethics by design or general phrases about responsible AI. It must demand material conditions for truth and accountability inside the very architecture of violence. Otherwise the next wars will be not only faster, but also less recoverable for memory, law and justice. If the fourth 68 is the threshold at which war enters the algorithm, then point 32 poses the next necessary condition: the algorithm cannot remain an opaque combat power. Otherwise history will receive not just a faster war, but a war without verifiable truth about itself.

33. War as a training corpus - how the fourth 68 turns the battlefield into a training set for the next war

Up to this point, the article traced how the four 68-thresholds coincide with new war accelerators, with new institutional needs, with new forms of vulnerability and with an increasingly dangerous compression of the time between perception and strike. But beneath this there is yet another, deeper mechanism without which the fourth 68 cannot be fully understood. War no longer merely uses AI. War is beginning to train the next war. This is qualitatively different from anything classical military history has known as "lessons from the battlefield." Now the battlefield is not only a place of experience. It is a source of annotated data, behavioral patterns, target classes, failure modes, environmental noise, electronic interference signatures and tactical adaptation traces, which then return back into the models, systems and doctrines.

It is precisely here that the fourth 68 reveals its deepest novelty. Previous thresholds change war through a new weapon, a new industry or a new limit of destruction. The current threshold begins to change war also through a new epistemic economy. Data from war become a strategic asset. Not only oil, metal, energy and the industrial base are valuable. Also valuable become the millions of observations from real combat flights, the movement models, the reactions under jamming, the ways in which real people, machines and infrastructures look and behave under strike. This means that part of the value of today's war is transferred forward not only as territory, victory or defeat, but as a training corpus for the next combat system.

Therefore the war in Ukraine must be thought of not only as a theater, but as the first major open training corpus of algorithmic warfare. This is not a metaphor. When millions of annotated images from combat flights are provided for training of allied models, and the goal is autonomous and AI-driven drone technologies, then the conflict is already more than a conflict. It is a training environment. From this moment on, a war begins to have a second life - after the combat action, in the next cycle of model training, testing, fine-tuning and operational deployment. This is a new form of historical transferability of violence.

It is important here to distinguish two things. Wars have always been a source of doctrinal lessons. Generals, staffs and armies have always observed, analyzed and adapted. But in the epoch of the fourth 68, adaptation is no longer only human and textual. It is machine-based and scalable. The difference is enormous. Human lessons are slow, selective and dependent on institutional memory. Machine-learnable combat data can be transferred, combined, multiplied and injected into new systems with much greater speed. This creates a feedback loop that did not exist in previous epochs in the same form: war produces data; data trains the model; the model accelerates the next operations; the next operations produce even more valuable data.

This new logic can be formulated as the data flywheel of war. The longer and more intensive the conflict, the more real data accumulates. The more data accumulates, the better the models for recognition, prioritization, navigation, adaptation under jamming and prediction of combat behavior become. The better the models become, the more effective the next cycles of war can be. And so the next war no longer starts from zero, but from the data inheritance produced by the previous one. If this logic remains unregulated, violence begins to generate not only destruction, but capitalizable knowledge for the next violence.

This leads to a new, deeply unpleasant political question. If war is a dataset, what incentives begin to emerge around it. It is not necessary for someone to cynically "want war" for the problem to appear. It is sufficient for the prolonged conflict to begin to look like a unique environment for collecting rare, high-value combat data. This creates a distortion. From a strategic perspective, certain actors may begin to see in prolonged war not only a cost, but also a source of capability building. This is a radically new danger. When the battlefield becomes simultaneously a tragedy and an infrastructure for training, the incentives around its termination can become more complex, more grim and less transparent.

From this it follows that the question of combat data is not a technical detail. It is a question of sovereignty, law and civilizational ethics. Whose are the combat data. Who has the right to collect, annotate, transfer, use for retraining, share with allies, integrate into new models and capitalize in subsequent weapons systems. If these questions remain unsettled, the fourth 68 will produce a new field of opaque power - not only power over the weapon, but power over the very training of the future weapon.

This necessitates the introduction of a new concept: sovereignty of combat data. It must mean that data extracted from real combat operations cannot be treated as a free and politically neutral resource. They must fall under a special regime, because they contain not only information about tactics, but also data about people, infrastructures, patterns of life, behavioral profiles, reactions under stress, electronic vulnerability and targets. Therefore, if combat data are used for training military AI, societies must know under what rules this happens, who controls the access, who performs audit, what is anonymized, what is not admitted and what is considered an impermissible transfer.

From here arise four concrete requirements for a future order.

First, a regime of battlefield data governance is needed. This means legal and institutional rules for the collection, annotation, storage, transfer and use of combat data for AI training. Without this, the training corpus of war will remain outside the legal framework, and the most critical layer of the next weapon will be regulated more weakly than the weapon itself.

Second, a regime of traceability for military training data is needed. If a given model is trained on combat data, it must be possible to trace on what type of corpus, from which theater, with what limitations, with what bias risks and with what safeguards it was trained. Without data lineage in the training pipeline, no one will know whether the next AI system carries within itself invalid, one-sided or outright dangerous historical distortions.

Third, a regime of limits on person-pattern training is needed. If military AI is trained on behavioral models of people, automobiles, buildings, settlements and patterns of life, then a particularly dangerous zone of fusion between intelligence, profiling and lethal use arises. This is the point at which the protection of human dignity and international humanitarian law begin to collide not only with the weapon, but with the training layer of the weapon.

Fourth, a regime against military monetization of combat data without a clear political mandate and public accountability is needed. If companies, states and alliances begin to treat prolonged war as a source of defensible model advantage, then a new moral and strategic deformation arises. Not because this automatically creates malice, but because it makes violence itself structurally useful for the next cycle of capability growth.

It is precisely here that point 33 connects with the entire preceding article. In points 20 and 21, the weapon was described as an accelerator. In points 28 and 29, it was shown how to observe it and whether it remains correctable. In points 30 and 32, the right to deceleration and the right to verifiability appeared. But if we miss point 33, the deepest movement will remain hidden: that the fourth 68 not only accelerates war in the present, but begins to turn war into an infrastructure for training future violence.

The shortest and hardest conclusion is the following. In the previous thresholds, humanity has often understood the new weapon only after it has already shown its destructive force. In the fourth 68, the risk is greater, because today's war not only inflicts destruction but simultaneously produces the learning material for the next one. If this process is not limited, civilization will enter a regime in which every major war leaves behind not only trauma and ruins, but an improved model for the next, even faster and more adaptive war. This is one of the darkest but also most useful truths that the deep reading of the fourth 68 allows us to see in time.

34. Compute as the new fissile material - the computational sovereignty of war

Up to this point, the text named the fourth 68 as the threshold of algorithmic warfare, of the accelerated kill chain, of combat data as a training corpus, of critical infrastructure as a field of vulnerability and of the right to deceleration and verifiability as new conditions for legitimacy. But beneath all these layers there is yet another material substrate without which almost nothing described can unfold at full scale. This is compute - the computational power capable of training, fine-tuning, executing and operationally maintaining large AI systems. In the epoch of the fourth 68, compute ceases to be a neutral technical prerequisite. It begins to transform into an independent strategic resource.

This requires a change in the way we think about military power. In the industrial age, steel, fuel and production capacity determined the tempo of war. In the nuclear age, fissile material and delivery systems determined the limits of destruction. In the algorithmic age, compute begins to play a similar role, but through a different mechanism. It is not the weapon itself. It is the carrier of the possibility for fast training, fast adaptation, mass simulation, high-frequency recognition, target generation, autonomous control loops and synchronization between data, models and platforms. If this carrier is concentrated, limited and strategically controlled, then access to it becomes a question of sovereignty.

This is precisely why the formulation "compute as the new fissile material" is not a metaphor for effect. It is an attempt to capture a structural resemblance. In both nuclear material and advanced compute, we see several common properties: strong concentration in a small number of actors and infrastructures, high cost of accumulation, difficulty of substitution, temptation for control over access and disproportionate strategic significance relative to its physical volume. The difference is that compute does not produce the final explosion directly. It produces the final acceleration. And in the epoch of the fourth 68, it is precisely the acceleration that begins to be more dangerous than the individual platform, because it compresses the time for human correction.

For this claim to be scientific, compute must be examined in its own triple form. First, as training - training compute, without which frontier models and complex military models cannot be created and brought to an operationally useful state. Second, as execution - inference compute, without which systems cannot operate in real time on the battlefield or in the command chains. Third, as infrastructure - datacenters, cloud regions, accelerators, interconnects, cooling, power stability and latency environment, without which training and inference remain limited, expensive or fragile. If any of these three layers is disrupted, military AI does not disappear entirely, but loses tempo, scale and resilience.

From here follows the first major conclusion of this point: data and models by themselves are not sufficient. Without compute, military AI remains a limited experiment. With compute, it transforms into a systemic capability. It is precisely here that the greatest misunderstanding in the public conversation lies. Usually "AI weaponization" is discussed as though the key is only the algorithm. But the algorithm without access to sufficient computational power cannot be trained on large-scale combat behavior, nor adapted in reasonable cycles, nor maintained in a complex multi-source targeting workflow in real time. Therefore the true question is not only who possesses the model. The true question is who possesses the sustainable, scalable and defensible compute upon which the model can become a military force.

This is already visible in policy as well. In 2026, the American debate over export controls showed that the problem is not only the export of advanced semiconductors as physical goods. Coming to the fore was also remote access to U.S.-hosted compute, including cloud environments that would allow foreign actors to circumvent part of the classical chip restrictions. Even when specific regulatory rules change, are delayed or withdrawn, the very fact that regulation is already thinking of remote compute access as a subject of national security is historically important. This means that compute has already crossed the threshold from a market resource to a strategic resource.

Here the second major line also appears: compute is no longer just a resource, but also a target. If AI, cloud and datacenters are part of the supporting layer of the next war, then they inevitably begin to appear within the perimeter of military risk as well. The Middle Eastern escalation of 2026 already showed warnings and reports according to which major technology companies, their cloud infrastructures and data centers in the Gulf can be thought of as admissible or semi-admissible targets in an expanding conflict. This is an extremely important change. War begins to move toward the very computational environment, not only toward the weapons it supports. From this moment on, compute is not simply a lever of power. It is also a fragile node of vulnerability.

Therefore computational sovereignty becomes a new dimension of defense. But it must not be understood superficially as "to have our own chips." This is only one part. True computational sovereignty includes at least six components: access to a sufficiently high class of hardware; a trusted cloud environment; energy and cooling resilience; protection of physical sites; legal and contractual control over remote access and priorities of use; and the ability to verify upon what, where and in what regime military models are trained or executed. A state or alliance that has models and data but lacks these six conditions does not possess real computational sovereignty. It possesses a dependent capability layer that can be disrupted or politically extorted.

This also changes the question of inequality in war. In classical models, the richer state has more weapons. In the new regime, the deeper asymmetry may be different: one side has not simply more weapons, but a faster training cycle, faster retraining, a larger simulation budget, higher inference throughput and scaling capacity. Thus inequality shifts from inventory advantage to compute advantage. And compute advantage can be less noticeable, harder to measure and harder to understand politically than the classical military balance. This is precisely why it is so dangerous: it can change war deeply before societies have found the language to describe the change.

Here follows the third major consequence: if compute is the new carrier of acceleration, then its management must also become a subject of law, inspection and an allied framework. It is not sufficient to regulate the final weapon if its computational foundation remains opaque, freely transferable and outside special oversight. From here arise several concrete proposals.

First, a regime of military-grade compute licensing is needed. This means that access to cloud or local compute above certain thresholds, when intended for military AI training, targeting, autonomous control or strategic inference, must be subject to a special regime of authorization, reporting and verification. The goal here is not to stop every military use of AI, but to make visible where and at what scale it is consolidating.

Second, a regime of metering and logging for high-risk compute is needed. If point 32 insisted on combat logs, point 34 insists on computational logs. This means a traceable record of training and operationally significant compute runs: on what hardware they occurred, in which environment, for what class of task, with what models, with what input corpora and in what authorization regime. Without compute logs, the trail of military AI will remain incomplete.

Third, a regime of remote attestation and secure provenance for military compute is needed. If a given model is operationally deployed or retrained for high-risk military use, it must be possible to attest in what computational environment this occurred, with what safeguards, under what policy layer and in what trusted execution context. This is especially important in a world where remote cloud access can be as strategically significant as the physical export of chips.

Fourth, a regime of allied compute compacts is needed. If alliances want to remain real and not merely political, they must begin to think of compute as a shared strategic resource. This means joint rules for access, reservation of capacities during crisis, allocation of priorities between civilian and military layers, common standards for logging and auditability and coordination on the protection of cloud and datacenter infrastructure. Without this, alliance cohesion will remain slower than the technical environment it must sustain.

Fifth, a mechanism for compute pause triggers is needed. Just as point 30 proposed the right to deceleration, here its material equivalent appears. In certain classes of escalation, autonomous incidents, AI-enabled targeting failures, large-scale infrastructure attacks or nuclear-adjacent crises, a legal possibility must exist to temporarily limit or place under special regime the use of certain categories of military compute. This sounds radical, but without such a mechanism the right to deceleration remains abstract. Deceleration must also have a material carrier. That carrier is compute.

Sixth, a principle of demilitarized compute zones is needed. Not the entire cloud can and should become a military reserve. Certain layers of socially significant, scientific, health-related, civilian and basic communications computational environment must remain outside military repurposing, except under extreme and clearly regulated regimes. Otherwise every major crisis will begin to pull civilizational compute toward military urgency and will strip the remaining systems bare.

Here the true radicalism of this point is visible. The computational sovereignty of war is not merely a technical topic for chip and cloud specialists. It is a new civilizational axis. Whoever holds compute holds the tempo of training. Whoever holds the tempo of training holds the speed of adaptation. Whoever holds the speed of adaptation begins to hold the tempo of war itself. In this sense, compute is not just infrastructure. It is a new form of latent military power.

According to "History of the Future," the deepest conclusion is the following: the fourth 68 not only enters the algorithm, not only turns war into a training corpus and not only requires the right to deceleration and verifiability. It displaces part of sovereignty itself into computational power. If civilization does not recognize this in time, it will continue to regulate the weapons while missing the material that makes the next weapons ever faster, more adaptive and more dependent on narrow, fragile and potentially targetable nodes. And then the historical mistake will not only be that we underestimated AI. It will be that we did not see compute as its true strategic carrier.

Sources

1) Encyclopaedia Britannica Editors - World War I - primary reference text for the beginning of the war, the participants and the general framework of the conflict.
https://www.britannica.com/event/World-War-I

2) Encyclopaedia Britannica Editors - Timeline of World War I - used for fixing 28.07.1914 as the start date of the war.
https://www.britannica.com/list/timeline-of-world-war-i

3) Encyclopaedia Britannica Editors - World War II - primary reference text for the beginning of the war in Europe on 01.09.1939 and the general framework of the conflict.
https://www.britannica.com/event/World-War-II

4) Encyclopaedia Britannica Editors - World War II: Human and Material Cost - used for the estimates of casualties and material cost of World War II.
https://www.britannica.com/event/World-War-II/Human-and-material-cost

5) Encyclopaedia Britannica Editors - Russia-Ukraine War - used for fixing 24.02.2022 as the start of the full-scale invasion and for the general military framework.
https://www.britannica.com/event/Russia-Ukraine-War

6) Encyclopaedia Britannica Editors - When did the Russia-Ukraine War begin? - used for a clear distinction between the beginning of the broader war in 2014 and the full-scale phase in 2022.
https://www.britannica.com/question/When-did-the-Russia-Ukraine-War-begin

7) Encyclopaedia Britannica Editors - Italo-Turkish War - used for the start date 29.09.1911, the geographic scope and the historical significance of the conflict.
https://www.britannica.com/event/Italo-Turkish-War

8) Rachel Simon - Italo-Turkish War 1911-1912 - an academic encyclopedic text on the causes, setting and significance of the war in Ottoman Libya.
https://encyclopedia.1914-1918-online.net/article/italo-turkish-war-1911-1912/

9) Encyclopaedia Britannica Editors - Military Aircraft - used for the claim that the first use of an airplane in war occurred in the Italo-Turkish War in 1911.
https://www.britannica.com/technology/military-aircraft

10) Richard C. Hall - Balkan Wars 1912-1913 - used for the start date of the First Balkan War and for the context of the Balkan escalation.
https://encyclopedia.1914-1918-online.net/article/balkan-wars-1912-1913/

11) Encyclopaedia Britannica Editors - Balkan Wars - used for the context and start date of the Second Balkan War.
https://www.britannica.com/topic/Balkan-Wars

12) Encyclopaedia Britannica Editors - War of 1812 - used for the start date 18.06.1812 and the brief characterization of the conflict.
https://www.britannica.com/event/War-of-1812

13) Encyclopaedia Britannica Editors - Crimean War Facts - used for the start date 04.10.1853 and the basic parameters of the Crimean War.
https://www.britannica.com/facts/Crimean-War

14) Encyclopaedia Britannica Editors - Serbo-Bulgarian War - used for the start date 14.11.1885 and the historical significance of the conflict.
https://www.britannica.com/event/Serbo-Bulgarian-War

15) Encyclopaedia Britannica Editors - Six-Day War - used for the start date 05.06.1967 and the strategic significance of the war.
https://www.britannica.com/event/Six-Day-War

16) Encyclopaedia Britannica Editors - Falkland Islands War - used for the start date 02.04.1982 and the general characterization of the conflict.
https://www.britannica.com/event/Falkland-Islands-War

17) Encyclopaedia Britannica Editors - Persian Gulf War - used for the start date 16.01.1991 and the framework of the operation against Iraq.
https://www.britannica.com/event/Persian-Gulf-War

18) Correlates of War Project - COW War Data, 1816-2007 - used as a methodological reference for what a formalized corpus of wars means and what the boundaries of modern war datasets are.
https://correlatesofwar.org/data-sets/cow-war/

19) Uppsala Conflict Data Program - UCDP Dataset Download Center - used as a methodological reference for start and end dates of conflict episodes and for the discipline of precise dating.
https://ucdp.uu.se/downloads/

20) UCDP/PRIO Armed Conflict Dataset Codebook - used for the principle that a startdate should be clearly defined and coded according to transparent rules.
https://ucdp.uu.se/downloads/ucdpprio/ucdp-prio-acd-172.pdf

21) Getty Vocabulary Editorial Guidelines - Appendix B: Dates and Date Authority - used for the principle of normalization to the proleptic Gregorian calendar when comparing historical dates.
https://www.getty.edu/publications/vocabularies-editorial-guidelines/ulan-guidelines/4_appendices/4.2/

22) World Bank Group - Ukraine: Fifth Rapid Damage and Needs Assessment - used for the assessment of damages and recovery needs after the full-scale Russian invasion.
https://www.worldbank.org/ext/en/country/ukraine

23) UNDP / RDNA5 - Ukraine Fifth Rapid Damage and Needs Assessment - additional source for infrastructure damages, socio-economic losses and the scale of recovery.
https://www.undp.org/ukraine/publications/ukraine-fifth-rapid-damage-and-needs-assessment-rdna5-february-2022-december-2025

24) Encyclopaedia Britannica Editors - World War I 1914-1918 - used for the duration and approximate scales of casualties of World War I.
https://www.britannica.com/discover/World-War-I

25) Reuters - Trump says U.S. will hit Iran "very hard" after easing sanctions on Russian oil - used as a source for the current USA-Israel-Iran war, the regional escalation, casualties and the systemic energy risk.
https://www.reuters.com/world/middle-east/both-sides-dig-iran-war-approaches-two-week-mark-2026-03-13/

26) Reuters - Analysts reassess oil price estimates as Iran conflict disrupts markets - used as a source for the oil shocks, the risk around the Strait of Hormuz and the transfer of the regional war to global economic consequences.
https://www.reuters.com/business/energy/analysts-reassess-oil-price-estimates-iran-conflict-disrupts-markets-2026-03-13/

27) Reuters - Trump advisers discussing options for acquiring Greenland, U.S. military is always an option - used as a source for the Greenland case, the shock to allied trust and tensions within NATO.
https://www.reuters.com/world/trump-advisers-discussing-options-acquiring-greenland-us-military-is-always-an-2026-01-06/

28) Reuters - From Greenland to Ukraine, Trump's centralized diplomacy creates whiplash for allies - used as a source for the allied nervousness, diplomatic unpredictability and the shaking of trust between the USA and European partners.
https://www.reuters.com/world/europe/greenland-ukraine-trumps-centralized-diplomacy-creates-whiplash-allies-2026-01-24/

29) Reuters - NATO will become more European-led, NATO Chief Rutte says - used as a source for the tendency toward a more European-led NATO and the rearrangement of roles within the alliance.
https://www.reuters.com/world/nato-will-become-more-european-led-nato-chief-rutte-says-2026-02-13/

30) Reuters - Europe needs European defence union, EU defence commissioner says - used as a source for European defense autonomy, doubt in old guarantees and the structural pressure for a new European defense architecture.
https://www.reuters.com/world/europe-needs-european-defence-union-eu-defence-commissioner-says-2026-02-10/

31) Reuters - Guyana asks World Court to block Venezuela's Esequibo election plans - used as a source for the ongoing Venezuela-Guyana tension around Essequibo and the risk of a resource territorial dispute becoming a broader strategic test.
https://www.reuters.com/world/americas/guyana-asks-world-court-block-venezuelas-esequibo-election-plans-2025-03-06/

32) Reuters - What is the border dispute between Venezuela and Guyana? - used as an additional source for the historical and geo-economic context of the Essequibo dispute.
https://www.reuters.com/world/americas/what-is-border-dispute-between-venezuela-guyana-2023-12-06/

33) Reuters - Ukraine opens battlefield data access to allies' AI models - used as a primary source for combat data from Ukraine, AI model training, autonomous systems and AI-drone acceleration as a new military regime.
https://www.reuters.com/business/aerospace-defense/ukraine-opens-battlefield-data-access-allies-ai-models-2026-03-12/

34) Reuters - Austria calls for rapid regulation as it hosts meeting on "killer robots" - used as an additional source that AI is already being used on the battlefield, including in drones in Ukraine under signal jamming.
https://www.reuters.com/technology/austria-calls-rapid-regulation-it-hosts-meeting-killer-robots-2024-04-29/

35) Reuters - U.S. looking at report that Israel used AI to identify bombing targets in Gaza - used as a cautious source for the public data regarding AI-assisted target identification by the Israeli army, without absolutizing this for every individual case.
https://www.reuters.com/world/middle-east/us-looking-report-that-israel-used-ai-identify-bombing-targets-gaza-2024-04-04/

36) Reuters - Bombed Iranian girls' school had vivid website, yearslong online presence - used not as proof of AI-centrality, but as a source for the limitation of our knowledge regarding the specific mechanism of target selection in the current war against Iran.
https://www.reuters.com/investigations/bombed-iranian-girls-school-had-vivid-website-yearslong-online-presence-2026-03-12/

37) Reuters - Was the U.S. capture of Venezuela's president legal? - used as a source for the very fact of the capture of Maduro and as grounds not to claim without evidence that the case was AI-driven.
https://www.reuters.com/world/us/was-us-capture-venezuelas-president-legal-2026-01-03/

38) Encyclopaedia Britannica Editors - Military aircraft - used as a source for the first use of an airplane in war and the first aerial bombardment in the Italo-Turkish War.
https://www.britannica.com/technology/military-aircraft

39) Encyclopaedia Britannica Editors - Bomber - used as an additional source for tracing the origin of aerial bombing to the Italo-Turkish War.
https://www.britannica.com/technology/bomber-aircraft

40) Encyclopaedia Britannica Editors - Weapons of World War I - used as a source for the package of accelerators in World War I - chemical weapons, aviation, submarine warfare and other key means.
https://www.britannica.com/list/weapons-of-world-war-i

41) Encyclopaedia Britannica Editors - World War II: Hiroshima and Nagasaki - used as a source for the first combat use of atomic weapons and the nuclear threshold introduced by the third 68 case.
https://www.britannica.com/event/World-War-II/Hiroshima-and-Nagasaki

42) Encyclopaedia Britannica Editors - Atomic bombings of Hiroshima and Nagasaki - used as an additional source for the atomic bombings as the first use of nuclear weapons in war.
https://www.britannica.com/event/atomic-bombings-of-Hiroshima-and-Nagasaki

43) Encyclopaedia Britannica Editors - League of Nations - reference text for the origin of the League of Nations after World War I and the principle of collective security.
https://www.britannica.com/topic/League-of-Nations

44) Encyclopaedia Britannica Editors - Geneva Gas Protocol - reference text for the 1925 protocol prohibiting the use of chemical and biological weapons in war.
https://www.britannica.com/event/Geneva-Gas-Protocol

45) United Nations - History of the United Nations - official history of the creation of the UN after World War II and the entry into force of the Charter of the United Nations.
https://www.un.org/en/about-us/history-of-the-un

46) United Nations - Predecessor: The League of Nations - official text on the League of Nations as the predecessor of the UN and the connection between the post-war institutional attempts of 1919 and 1945
https://www.un.org/en/about-us/history-of-the-un/predecessor

47) United Nations - Preparatory Years: UN Charter History - official text on the signing of the Charter of the United Nations and the preparation of the post-war international architecture.
https://www.un.org/en/about-us/history-of-the-un/preparatory-years

48) United States Holocaust Memorial Museum - The Nuremberg Trials - encyclopedic text on the Nuremberg Trials as a post-war legal and moral corrective after World War II.
https://encyclopedia.ushmm.org/content/en/article/the-nuremberg-trials

49) International Committee of the Red Cross - Customary International Humanitarian Law - summarizing text on the Geneva Conventions of 1949 as the foundation of contemporary international humanitarian law.
https://www.icrc.org/sites/default/files/external/doc/en/assets/files/other/customary-international-humanitarian-law-i-icrc-eng.pdf

50) International Atomic Energy Agency - The IAEA and the Non-Proliferation Treaty - official text on the NPT as the central regime for limiting the proliferation of nuclear weapons.
https://www.iaea.org/topics/non-proliferation-treaty

51) Encyclopaedia Britannica Editors - Italo-Turkish War - reference text for the start of the Italo-Turkish War on 29 September 1911 and its place in the pre-war order of Europe.
https://www.britannica.com/event/Italo-Turkish-War

52) Encyclopaedia Britannica Editors - Timeline of World War I - chronological reference in which 28 July 1914 is established as the beginning of World War I.
https://www.britannica.com/list/timeline-of-world-war-i

53) Encyclopaedia Britannica Editors - World War II: The war in Europe, 1939-41 - reference text for the German attack on Poland on 1 September 1939 as the beginning of the war in Europe.
https://www.britannica.com/event/World-War-II/The-war-in-Europe-1939-41

54) UK Parliament House of Commons Library - Conflict in Ukraine: A timeline (current conflict, 2022 - present) - chronological reference that establishes 24 February 2022 as the beginning of the current full-scale phase of the war.
https://commonslibrary.parliament.uk/research-briefings/cbp-9847/

55) Spyros Makridakis, Robin M. Hogarth, Anil Gaba - Forecasting and Uncertainty: A Survey - a survey academic text on the limits of forecasting and on the fact that under high uncertainty it is more honest to work with corridors and warnings, not with false precision.
https://www.acg.edu/ckeditor_assets/attachments/1893/forecasting_and_uncertainty.pdf

56) J. Scott Armstrong, Kesten C. Green, Andreas Graefe - Causal Forces: Structuring Knowledge for Time-Series Extrapolation - academic text on careful extrapolation of series and the need to damp trends when the causal forces are unclear or variable.
https://faculty.wharton.upenn.edu/wp-content/uploads/1993/01/144-JSA-Causal-Forces-Structuring-Knowledge-for-Time-series-Extrapolation.pdf

57) United Nations Office for Disarmament Affairs - Lethal Autonomous Weapon Systems - official overview of autonomous weapons systems, the UN position and the call for a binding international instrument by 2026
https://disarmament.unoda.org/en/our-work/emerging-challenges/lethal-autonomous-weapon-systems

58) International Committee of the Red Cross - Preserving human control over the use of force: A call to regulate lethal autonomous weapon systems under international law - official ICRC position on prohibitions and limitations on autonomous weapons and on preserving human control over the use of force.
https://www.icrc.org/en/statement/preserving-human-control-over-use-force-call-regulate-lethal-autonomous-weapon-systems

59) Council of Europe - The Framework Convention on Artificial Intelligence - official text on the Council of Europe framework convention, including its scope and limitations regarding defense and national security.
https://www.coe.int/en/web/artificial-intelligence/the-framework-convention-on-artificial-intelligence

60) European Union AI Act Service Desk - Recital 24 - official clarification that military, defense and national security applications of AI are outside the scope of the AI Act, which outlines a real regulatory gap.
https://ai-act-service-desk.ec.europa.eu/en/ai-act/recital-24

61) NATO - Resilience, civil preparedness and Article 3 - official text on resilience, access to critical infrastructure and civil preparedness as central elements of collective security.
https://www.nato.int/en/what-we-do/deterrence-and-defence/resilience-civil-preparedness-and-article-3

62) European Commission - Critical infrastructure resilience at EU-level - official overview of the critical entities resilience regime, CER Directive, national strategies, risk assessments and coordination at EU level.
https://home-affairs.ec.europa.eu/policies/internal-security/counter-terrorism-and-radicalisation/protection/critical-infrastructure-resilience-eu-level_en

63) European Commission - White Paper for European Defence - Readiness 2030 - official document on defense readiness, strategic reserves, industrial readiness pools, supply security and the accelerated European defense industry.
https://commission.europa.eu/document/download/e6d5db69-e0ab-4bec-9dc0-3867b4373019_en

64) International Committee of the Red Cross - Autonomous Weapons - official thematic overview of the legal, ethical and technological questions around autonomous weapons and meaningful human control.
https://www.icrc.org/en/law-and-policy/autonomous-weapons

65) OECD - Enhancing the resilience of communication networks - report on the resilience of communication networks, including the role of undersea cables, the frequency of incidents and the systemic risk to digital connectivity.
https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/05/enhancing-the-resilience-of-communication-networks_a47d78a1/d6920477-en.pdf

66) European Commission - Report on Security and Resilience of EU Submarine Cable Infrastructures - official report on mapping, risk, stress tests and vulnerabilities of European undersea cable infrastructure.
https://digital-strategy.ec.europa.eu/en/library/report-security-and-resilience-eu-submarine-cable-infrastructures

67) European Commission and NATO - EU-NATO Final Assessment Report on the Resilience of Critical Infrastructure - official joint report on energy, transport, digital infrastructure and space as critical layers of contemporary security.
https://commission.europa.eu/system/files/2023-06/EU-NATO_Final%20Assessment%20Report%20Digital.pdf

68) International Energy Agency - Global Critical Minerals Outlook 2025 - official analysis on critical minerals, the concentration of supplies and the strategic significance of mineral security.
https://www.iea.org/reports/global-critical-minerals-outlook-2025

69) OECD - Vulnerabilities in the Semiconductor Supply Chain - analytical report on the concentrations, dependencies and systemic vulnerabilities in the semiconductor supply chain.
https://www.oecd.org/content/dam/oecd/en/publications/reports/2023/06/vulnerabilities-in-the-semiconductor-supply-chain_f4de7491/6bed616f-en.pdf

70) European Commission - European Chips Act: Security of Supply and Resilience - official explanation of the European framework for semiconductor supply resilience and reduction of external dependencies.
https://digital-strategy.ec.europa.eu/en/factpages/european-chips-act-security-supply-and-resilience

71) International AI Safety Report 2026 - international scientific report on AI risks, including deepfakes, the erosion of trust and the undermining of the reliability of evidence.
https://internationalaisafetyreport.org/publication/international-ai-safety-report-2026

72) World Economic Forum - The Global Risks Report 2026 - global risk report on the role of disinformation, cyber risk, conflicts and social polarization as systemic short-term threats.
https://reports.weforum.org/docs/WEF_Global_Risks_Report_2026.pdf

73) International Telecommunication Union - International Advisory Body for Submarine Cable Resilience - official announcement of the creation of an international advisory body for the resilience of undersea cables.
https://www.itu.int/en/mediacentre/Pages/PR-2024-11-29-advisory-body-submarine-cable-resilience.aspx

74) Ian Duncan - Anthropic's AI tool Claude central to U.S. campaign in Iran - Washington Post report describing the U.S. military use of Palantir's Maven Smart System with Anthropic's Claude in the Iran campaign, including the role of the system in rapidly generating, locating and prioritizing targets.
https://www.washingtonpost.com/technology/2026/03/04/anthropic-ai-iran-campaign/

75) Financial Times - The AI-driven 'kill chain' transforming how the US wages war - Financial Times analysis of AI in the U.S. military targeting cycle, including Palantir Maven, Anthropic and the acceleration of planning, selection and assessment of strikes in Iran.
https://www.ft.com/content/fedb262e-e6db-40bc-a4d0-080812f0f82b

76) Reuters - Pentagon informed Anthropic it is a supply-chain risk, official says - Reuters report confirming the Pentagon designation, the continued relevance of Claude in defense workflows, and the description of Maven as a system used for intelligence analysis and weapons targeting in the context of Iran.
https://www.reuters.com/technology/pentagon-informed-anthropic-it-is-supply-chain-risk-official-says-2026-03-05/

77) Reuters - Anthropic has strong case against Pentagon blacklisting, legal experts say - Reuters report describing the legal dispute, the government's continued prior use of Claude, and the broader military significance of Anthropic's tools in recent operations.
https://www.reuters.com/legal/legalindustry/anthropic-has-strong-case-against-pentagon-blacklisting-legal-experts-say-2026-03-11/

78) WIRED - Palantir Demos Show How the Military Could Use AI Chatbots to Generate War Plans - report on Palantir's integration of Claude into military software such as Maven and the use of chat-based assistants for planning, surveillance interpretation and operational support.
https://www.wired.com/story/palantir-demos-show-how-the-military-can-use-ai-chatbots-to-generate-war-plans/

79) Financial Times - Iran and the rising perils of AI in warfare - Financial Times commentary on the battlefield role of AI in the Iran conflict, including machine-assisted decision support, target processing and the ethical risks of accelerated warfare.
https://www.ft.com/content/5d294db8-6917-4085-ba8a-3afbc439b999

80) Reuters - Ukraine opens battlefield data access to allies' AI models - report on the opening of Ukrainian combat data for training of allied AI models, autonomous systems and drone technologies.
https://www.reuters.com/business/aerospace-defense/ukraine-opens-battlefield-data-access-allies-ai-models-2026-03-12/

81) Reuters - Palantir faces challenge to remove Anthropic from Pentagon's AI software - report on the fact that Maven Smart System uses Claude and serves for intelligence analysis and weapons targeting.
https://www.reuters.com/technology/palantir-faces-challenge-remove-anthropic-pentagons-ai-software-2026-03-04/

82) Reuters - US allies rebuff Trump's request for support in Strait of Hormuz - report on the disagreements between allies regarding the conflict around Iran and Hormuz and the alliance tensions within the West.
https://www.reuters.com/business/aerospace-defense/us-allies-rebuff-trumps-request-support-strait-hormuz-2026-03-16/

83) Reuters - NATO allies will keep supporting Ukraine despite situation in Iran, Rutte says - report on NATO's attempt to simultaneously hold the Ukrainian and Middle Eastern pressure within one common allied framework.
https://www.reuters.com/world/nato-allies-will-keep-supporting-ukraine-despite-situation-iran-rutte-says-2026-03-05/

84) Reuters - Europe's main military powers to develop low-cost air-defence systems - report on the accelerated European transition to cheap autonomous and semi-autonomous counter-drone defense systems.
https://www.reuters.com/business/aerospace-defense/europes-strongest-military-powers-plan-drone-defence-programme-2026-02-20/

85) Reuters - NATO expected to launch Arctic Sentry mission in coming days - report on the new NATO mission in the Arctic against the backdrop of tensions around Greenland and the expansion of the crisis map.
https://www.reuters.com/world/nato-expected-launch-arctic-sentry-mission-coming-days-sources-say-2026-02-09/

86) International Committee of the Red Cross - Autonomous weapon systems and international humanitarian law: Selected issues - position paper on the need for an international instrument to prohibit unacceptable autonomous weapons and to limit the rest.
https://www.icrc.org/sites/default/files/2026-03/4896_002_Autonomous_Weapons_Systems_-_IHL-ICRC.pdf

87) United Nations Office for Disarmament Affairs - Lethal Autonomous Weapon Systems - official overview of the international debate and the Secretary-General's call for states to adopt by 2026 a binding instrument to prohibit systems without human control or oversight and to regulate the rest.
https://disarmament.unoda.org/en/our-work/emerging-challenges/lethal-autonomous-weapon-systems

88) International Committee of the Red Cross - Preserving human control over the use of force - official call for a new legally binding instrument with clear prohibitions and limitations on autonomous weapons and for preserving human control over decisions of life and death.
https://www.icrc.org/en/statement/preserving-human-control-over-use-force-call-regulate-lethal-autonomous-weapon-systems

89) International Committee of the Red Cross - Autonomous weapon systems and international humanitarian law: Selected issues - analytical document on the difficulty of predicting and limiting the effects of autonomous weapons and their connection to the principles of distinction, proportionality and precautions in attack.
https://www.icrc.org/sites/default/files/2026-03/4896_002_Autonomous_Weapons_Systems_-_IHL-ICRC.pdf

90) NATO - Resilience, civil preparedness and Article 3 - official text on resilience as the ability to prepare, withstand, respond and recover from shocks and disruptions, including with respect to critical infrastructure and hybrid or armed attacks.
https://www.nato.int/en/what-we-do/deterrence-and-defence/resilience-civil-preparedness-and-article-3

91) NATO - The Hague Summit Declaration - official text on the priority on critical infrastructure protection, civil preparedness, resilience, innovation and defence industrial base in the current strategic environment.
https://www.nato.int/en/about-us/official-texts-and-resources/official-texts/2025/06/25/the-hague-summit-declaration

92) NATO - NATO and Ukraine share lessons learned for strengthening civilian resilience and critical infrastructure - official text on the transfer of Ukrainian experience toward civilian resilience, crisis management and protection of critical infrastructure.
https://www.nato.int/en/news-and-events/articles/news/2026/01/14/nato-and-ukraine-share-lessons-learned-for-strengthening-civilian-resilience-and-critical-infrastructure

93) Council of the European Union - Hybrid threats - official overview of European measures against hybrid threats and the Critical Entities Resilience Directive as an instrument for reducing vulnerabilities and strengthening the physical resilience of critical entities.
https://www.consilium.europa.eu/en/policies/hybrid-threats/

94) United Nations Office for Disarmament Affairs - Responsible AI in the Military Domain - coverage of the international debate on the necessity for decisions on the use of nuclear weapons to remain under human control and on the risk of convergence between AI and the nuclear sphere.
https://disarmament.unoda.org/en/updates/responsible-design-conversation-global-commission-responsible-ai-military-domain

95) United Nations Office for Disarmament Affairs - Lethal Autonomous Weapon Systems - official overview of the international debate and the Secretary-General's call for states to adopt by 2026 a binding instrument to prohibit systems without human control or oversight and to regulate the rest.
https://disarmament.unoda.org/en/our-work/emerging-challenges/lethal-autonomous-weapon-systems

96) International Committee of the Red Cross - Limits on Autonomy in Weapon Systems - official ICRC position that preserving human control over the use of force requires clear limitations on autonomous weapons.
https://www.icrc.org/en/document/limits-autonomous-weapons

97) Jessica Dorsey - The erosion of human(e) judgement in targeting? Quantification logics, AI-enabled decision support systems and proportionality assessments in IHL - academic text on how AI-enabled decision support systems accelerate the observe-orient-decide-act cycle and can erode human judgement, precautions in attack and proportionality assessments.
https://international-review.icrc.org/sites/default/files/reviews-pdf/2026-02/the-erosion-of-humane-judgement-in-targeting-quantification-logics-ai-enabled-decision-support-systems-a.pdf

98) United Nations General Assembly - General Assembly Adopts Resolution on Possible Risks of Integration of Artificial Intelligence in Command, Control and Communications Systems of Nuclear Weapons - official text on the internationally recognized risk of AI in the nuclear command-and-control environment.
https://press.un.org/en/2025/ga12736.doc.htm

99) United Nations Office for Disarmament Affairs - A Conversation with the Global Commission on Responsible AI in the Military Domain - official text emphasizing the need for the decision on the use of nuclear weapons to remain under human control and the concern about the convergence between AI and the nuclear sphere.
https://disarmament.unoda.org/en/updates/responsible-design-conversation-global-commission-responsible-ai-military-domain

100) NATO - Resilience, civil preparedness and Article 3 - official text on resilience as the ability to prepare, withstand, respond and recover from major shocks, including from hybrid and armed threats.
https://www.nato.int/en/what-we-do/deterrence-and-defence/resilience-civil-preparedness-and-article-3

101) NATO - The Hague Summit Declaration - official text on critical infrastructure, resilience, innovation and defence industrial base as central elements of the current strategic environment.
https://www.nato.int/en/about-us/official-texts-and-resources/official-texts/2025/06/25/the-hague-summit-declaration

102) NATO - Strengthened Resilience Commitment - official text on national and collective resilience as a condition for security and for maintaining essential governmental and societal functions under pressure.
https://www.nato.int/en/about-us/official-texts-and-resources/official-texts/2021/06/14/strengthened-resilience-commitment

103) European Commission - Critical infrastructure resilience at EU-level - official overview of the Critical Entities Resilience Directive and the requirement for critical services to remain operational under natural and man-made disruptive incidents.
https://home-affairs.ec.europa.eu/policies/internal-security/counter-terrorism-and-radicalisation/protection/critical-infrastructure-resilience-eu-level_en

104) International Energy Agency - Energy System Resilience: Executive Summary - official analysis on why resilience must be placed at the center of energy planning and how power systems can avoid catastrophic societal impacts.
https://www.iea.org/reports/energy-system-resilience/executive-summary

105) World Health Organization - Health Systems Resilience - official text on the fact that resilient health systems must be able to prevent, prepare for, detect, adapt, respond to and recover from shocks while maintaining essential health services, including in fragile and conflict settings.
https://www.who.int/teams/primary-health-care/health-systems-resilience

106) International Committee of the Red Cross - Respecting and Protecting Health Care in Armed Conflicts and in Situations Not Covered by International Humanitarian Law - legal and operational text on the protection of medical units and access to healthcare in conditions of conflict.
https://www.icrc.org/sites/default/files/document/file_list/health-care-law-factsheet-icrc-eng.pdf

107) OECD - Enhancing the Resilience of Communication Networks - analytical report on the resilience of communication networks and the critical role of submarine cables, which carry over 99% of international IP data traffic.
https://www.oecd.org/content/dam/oecd/en/publications/reports/2025/05/enhancing-the-resilience-of-communication-networks_a47d78a1/d6920477-en.pdf

108) UNESCO - Building resilience: disaster risk management for documentary heritage and digital archives - official material on resilience of documentary heritage and digital collections and the need for archives and digital evidentiary collections to be protected during crisis and destruction.
https://www.unesco.org/en/articles/building-resilience-disaster-risk-management-documentary-heritage-and-digital-archives-training

109) United Nations Office for Disarmament Affairs - Artificial intelligence in the military domain - official UN overview of the main issues in military AI governance, including reliability, predictability, explainability, human judgment and accountability.
https://disarmament.unoda.org/en/our-work/emerging-challenges/artificial-intelligence-military-domain

110) International Committee of the Red Cross - Autonomous weapon systems and international humanitarian law: selected issues and perspectives from the ICRC - official legal text emphasizing that humans must determine the lawfulness of attacks and remain responsible for these decisions, which cannot be delegated to machine processes.
https://www.icrc.org/sites/default/files/2026-03/4896_002_Autonomous_Weapons_Systems_-_IHL-ICRC.pdf

111) United Nations - Governing AI for Humanity: Final Report - final report that includes receiving and investigating reports of incidents or misuses, reporting serious breaches, verifying compliance and coordinating accountability as core governance functions.
https://www.un.org/sites/un2.un.org/files/governing_ai_for_humanity_final_report_en.pdf

112) Jessica Dorsey - The erosion of human(e) judgement in targeting? Quantification logics, AI-enabled decision support systems and proportionality assessments in IHL - academic text on how AI-enabled decision support systems can erode human judgment, precautions in attack and proportionality review.
https://international-review.icrc.org/sites/default/files/reviews-pdf/2026-02/the-erosion-of-humane-judgement-in-targeting-quantification-logics-ai-enabled-decision-support-systems-a.pdf

113) National Institute of Standards and Technology - Artificial Intelligence Risk Management Framework (AI RMF 1.0) - framework for trustworthy AI, including transparency, traceability, accountability and documentation as risk-management requirements.
https://nvlpubs.nist.gov/nistpubs/ai/nist.ai.100-1.pdf

114) Coalition for Content Provenance and Authenticity - C2PA / Content Credentials - official standard and infrastructure for content provenance, transparency and authenticity, relevant to the evidentiary chain and the verifiability of digital content.
https://c2pa.org/

115) European Commission - Second draft Code of Practice on marking and labelling AI-generated content - official European line on provenance, marking, metadata and labelling of AI-generated content as mechanisms for transparency and traceability.
https://digital-strategy.ec.europa.eu/en/library/commission-publishes-second-draft-code-practice-marking-and-labelling-ai-generated-content

116) European Parliamentary Research Service - Information manipulation in the age of generative artificial intelligence - analytical text on synthetic media, provenance signals, watermarking and the challenge of preserving trust in evidence under generative AI conditions.
https://www.europarl.europa.eu/RegData/etudes/BRIE/2025/779259/EPRS_BRI%282025%29779259_EN.pdf

117) Reuters - Ukraine opens battlefield data access to allies' AI models - report on the decision of Ukraine to open access to combat data for allies and companies for the purpose of training AI-drone technologies, including millions of annotated images from real combat flights.
https://www.reuters.com/business/aerospace-defense/ukraine-opens-battlefield-data-access-allies-ai-models-2026-03-12/

118) European Parliament Research Service - Defence and artificial intelligence - analytical document that describes Russia's war against Ukraine as an "AI war lab" and examines the role of AI in geospatial intelligence, unmanned systems operations, military training and cyber warfare.
https://www.europarl.europa.eu/RegData/etudes/BRIE/2025/769580/EPRS_BRI%282025%29769580_EN.pdf

119) United Nations Office for Disarmament Affairs - Governance of Artificial Intelligence in the Military Domain - document emphasizing the importance of representative and contextual datasets for training and testing of AI-enabled military systems and the risks of bias and data governance gaps.
https://unodaweb.unoda.org/public/2024-06/OP42.pdf

120) International Committee of the Red Cross - Artificial Intelligence use for Military Purposes - legal and operational material that notes that military AI requires a sufficient quantity of data, as well as effective preparation and coding of training data before integration into the systems.
https://casebook.icrc.org/print/pdf/node/21289

121) United Nations General Assembly - Artificial intelligence in the military domain - report that explicitly addresses training data bias and the need for guidelines on responsible data governance specially tailored to military AI.
https://docs.un.org/en/A/80/78?direct=true

122) UNIDIR - The Global Prism of Military AI Governance - report that examines operationalization of responsible AI in the military domain and includes data practices, incident response and accountability pathways as separate governance axes.
https://unidir.org/wp-content/uploads/2026/02/UNIDIR_The_Global_Prism_of_Military_AI_Governance.pdf

123) CIGI / UNIDIR - Inclusion of Data Governance in Responsible Military AI - analytical text that formulates that data plays a critical role in the training, testing and use of artificial intelligence in the military domain and that data governance must be a central part of responsible military AI.
https://www.cigionline.org/documents/2834/Afina-Grand_Clement.pdf

124) International AI Safety Report 2026 - synthetic scientific report that explicitly states that pre-training of modern models typically requires large amounts of compute and data, and access to computational resources is sensitive to political and regulatory changes.
https://internationalaisafetyreport.org/sites/default/files/2026-02/international-ai-safety-report-2026.pdf

125) Reuters - U.S. Commerce Department withdraws planned rule on AI chip exports - report on the debate in the US administration on a new framework for controlling AI chips and access to them, including thresholds, bilateral deals and linking exports to security and investments in data centers.
https://www.reuters.com/business/us-commerce-department-withdraws-planned-rule-ai-chip-exports-government-website-2026-03-13/

126) Reuters - U.S. mulls new rules for AI chip exports, including requiring investments by foreign nations - report on the discussed new rules, which now think not only of the chips, but also of the broader infrastructure of advanced AI access as a matter of national security.
https://www.reuters.com/world/us-mulls-new-rules-ai-chip-exports-including-requiring-investments-by-foreign-2026-03-05/

127) Security and Technology Policy Accelerator - A Changing Export Control Landscape - analytical document that explicitly formulates that the restriction of remote access marks a transition from control over physical chips to managing the accumulation and use of advanced compute over time.
https://securityandtechnology.org/wp-content/uploads/2026/02/A-Changing-Export-Control-Landscape.pdf

128) WIRED - Iran Warns US Tech Firms Could Become Targets as War Expands - report on how in the escalation around Iran technology companies, cloud facilities and data centres in the Gulf begin to appear as real military vulnerabilities and potential targets.
https://www.wired.com/story/iran-warns-us-tech-firms-could-become-targets-as-war-expands

129) Euronews Next - Data centres are the new target in modern warfare during Iran war, experts say - analysis on the fact that data centres begin to be perceived as a new type of strategic target, because they support cloud services, banking apps and AI platforms.
https://www.euronews.com/next/2026/03/12/data-centres-are-the-new-target-in-modern-warfare-during-iran-war-experts-say

130) OECD AI Policy Observatory - Data centres - material on data centres as strategic digital infrastructure and sovereign cloud architectures, relevant to the idea of computational sovereignty.
https://oecd.ai/en/dashboards/policy-initiatives/data-centres