What kind of gamedo you thinkyou’re in?

A zero-sum game is a formal structure. In Theory of Games and Economic Behavior, John von Neumann and Oskar Morgenstern put it this way: if the sum of all payments received by all players is always zero, “the players pay only to each other, and that no production or destruction of goods is involved.”

They call that a zero-sum game. Where the total social product depends on how the players behave, and is in general neither zero nor constant, they call the game non-zero-sum.

In a two-person zero-sum game, they write, anything one player wins is necessarily lost by the other. That is an absolute antagonism of interests. A constant-sum game is the same picture with a fixed total that is not zero. One player’s gain is still the other’s loss relative to that constant. They treated the move from zero-sum to constant-sum as not a really significant extension, because the two are strategically equivalent.

Thomas Schelling, in 1958, treated the zero-sum game as a limiting case of pure conflict, not the place to start. The situations he wanted to understand mix conflict with mutual dependence: wars, strikes, negotiations, deterrence. Even there, mutual dependence is part of the logical structure and demands some kind of collaboration or mutual accommodation, “even if only in the avoidance of mutual disaster.”

He called the mix a mixed-motive game. The phrase names the relation between players, not a confused person.

So the word is already precise before anyone turns it into a worldview. Some games really do pay only from one player to another. Some do not. The question on this table is what happens when a rule that fits the first kind is used, half awake, for every kind.

Joanna Różycka-Tran, Paweł Boski, and Bogdan Wojciszke named a social axiom for that conviction: belief in a zero-sum game. In their 2015 abstract, it is a belief system about the antagonistic nature of social relations, “that one person’s gain is possible only at the expense of other persons.”

It can sit as a personal conviction and as a cultural worldview. Their project covered 37 nations. They report an affinity with societal cynicism, and they say the belief is moderated by cultural dimensions and by socio-economic indices. The abstract does not report an experiment. It does not diagnose a character.

Shai Davidai and Stephanie J. Tepper, reviewing the psychology in 2023, define zero-sum beliefs as subjective: “independent of the actual distribution of resources, one party’s gains are inevitably accrued at other parties’ expense.” People hold them about other people, about groups, and about nations. The review is not itself an experiment. It argues that three channels keep showing up: threat, real or imagined scarcity, and a lack of deliberation about what an exchange does over time.

Threat need not match the facts. Merely seeing a resource as scarce can foster the belief even when the resource is not limited. Unequal distributions, they note, are not inherently zero-sum.

Much of the literature is correlational, because the belief is hard to manipulate cleanly. Where studies do find causal effects, those effects belong to those studies, not to a verdict on a visitor. The review also says the beliefs are not uniformly negative. A correlation with distrust can run in more than one direction. A third thing, scarcity itself, can feed both.

John Duckitt’s dual-process work describes a related social worldview: the competitive jungle. In the 2002 formulation with colleagues, the construct begins as the belief that the social world is a competitive jungle “characterized by a ruthless, amoral struggle for resources and power in which might is right.”

One item in the scale they developed says “it’s a dog-eat-dog world where you have to be ruthless at times.” That is a measured belief about other people. This page will not pin it on you.

The quiet move is easy to miss. Someone must lose. There is never enough. People are rivals first. If I do not take it, somebody else will. None of these has to be spoken to do its work. A rule that is true of chips, or of a seat, becomes the background light in which money, status, credit, institutions, and the future are seen.

The belief can be accurate in one room and false in the next. The trouble starts when the rooms are no longer told apart.

The chips on the left were rival. The method on the right was not. The seat refused to multiply. That is the distinction a slogan about abundance usually smears.

Some things are finite or rivalrous. A harvest at its ecological limit. A bed in a full ward. An orbit slot. A vote one person cannot cast twice in the same race. Time in a day. If you take the last of these, someone else does not have it. Calling that zero-sum is not a mood. It is a description.

Other things are non-rival, reproducible, or generative. A proof, once found, can live in two minds. A tool can be built that neither person could finish alone. An institution can make an exchange possible that did not exist on Monday. Trade, at its plainest, is often not a transfer of a fixed pile. Each side can leave with something valued more than what was given up.

That can fail. Fraud exists. Power can set the terms. Costs can land on someone who was not in the room. None of that restores a constant sum. It means a larger game can still be played badly.

Von Neumann and Morgenstern already marked the economic point inside the definition. Where production is possible, the total is not fixed by the form of the game. It depends on what the players do. A positive-sum outcome is possible in the structure. It is not guaranteed. Cooperation can collapse. Incentives matter. Free riders exist. Scarcity does not vanish because someone would like a kinder story.

Paul Samuelson, in 1954, separated ordinary private goods, which can be parcelled out, from what he called collective consumption goods: goods “which all enjoy in common in the sense that each individual’s consumption of such a good leads to no subtraction from any other individual’s consumption of that good.” That sentence is about joint consumption. It does not, in the wording verified here, add the later textbook word non-excludable.

Mancur Olson used “common, collective, or public good” for a different cut. If one person in a group consumes it, it cannot feasibly be withheld from the others in that group. He was explicit that this is not Samuelson’s pure public good, because jointness of supply is not required for his definition.

His claim about action is stark, and it is not a sermon about selfish souls. Unless the group is quite small, or there is coercion or some other special device, “rational, self-interested individuals will not act to achieve their common or group interests.” In a large organization your effort may not change the result, and you can enjoy what others build whether or not you helped. Unions, he notes, already worried about the free rider.

Elinor Ostrom spent a career refusing the next shortcut: that whatever is not a neat market must be handed to a single government, or the reverse. In the 2009 Nobel lecture she describes research that does not fit a world divided into “the market” and “the state.” People build private, governmental, and community arrangements. Those arrangements produce productive and innovative outcomes, and also destructive and perverse ones.

In the laboratory, anonymous strangers with no way to talk overharvest a common-pool resource, as noncooperative models predict. Allowing them to talk can reduce the overharvest. That is not niceness as a policy. It is a change in the situation.

In the field, the Workshop’s comparisons did not crown one kind of owner. Of 47 irrigation systems in one analysis she reports, the government-managed systems less often showed high performance than the farmer-managed ones. The numbers are hers, and they are not a universal ranking: 12 government-managed, of which about 40 percent were high performing, and 25 farmer-managed, of which over 70 percent were.

She also reports inshore fisheries, a later Nepal irrigation set, and forests through a network of countries. Governing the Commons looks at mountain grazing and forests in Switzerland and Japan, and irrigation in Spain and the Philippine Islands. Earlier work on California groundwater and on metropolitan policing taught her that many centers of decision can be a system rather than chaos.

Polycentric, in the sense she takes from Vincent Ostrom, Tiebout, and Warren, means many centers that are formally independent. Whether they actually function as a system is an empirical question.

One-size-fits-all policies, she says, are not effective. Turning a resource over to a government, privatizing it, or decentralizing it can each fail. What works is closer to fit: boundaries people understand, rules matched to the place, and a real part in making those rules for the people who live with them. Institutions are not a mood of cooperation. They are how cooperation becomes possible when goodwill runs out.

Civics, here, is the relation between a person and the shared systems in which people act together. It is not a party and not a candidate. It is not the word government used as a synonym for everything public. Law, markets, commons, voluntary associations, science, infrastructure, schooling, and the ordinary keeping of promises all change what game is available.

A citizen meets a question the private pile does not ask. What do we owe to arrangements none of us can build alone, simply because we inhabit them with other people? A road. A court. A language for measurement. A rule that lets strangers trade without a private army. None of these is free of conflict. Building them is not the same act as redistributing a fixed pile, and it is not the same act as wishing people were kinder.

Agency, in this sense, is not only a choice among pieces that already exist. Knowledge can be discovered. A tool can be invented. An institution can be redesigned. A relationship can produce a capacity neither person had. What can exist tomorrow that does not exist today? History, material limits, and prior choices still bind the answer. A wider field of possibility is not a blank slate.

In the same stretch he says that if you have a zero-sum mindset, the only way to get ahead is by taking things from others. He says the pie is not fixed, and that it is much better to work on adding to the economic pie, creating more than you consume.

A little earlier he asks whether you are contributing more than you consume. He also says that people, including very smart people, can do things that seem morally questionable because they hold a zero-sum mindset at a “base, sort of axiomatic level,” and that they may not realize it consciously. He wants, in his words, “not a zero-sum mindset, or have more of a grow the pie mindset.”

This page does not hand him the last word. The founding texts already contain the formal point he is reaching for: when production is possible, the sum need not be fixed. The psychology already contains the warning: a person can use a zero-sum model without noticing, and the model can be false of the situation in front of them.

What he adds is a moral hinge, not a proof. If you believe the only way ahead is to take, you will try to take. That is an argument about motivation. It does not show that every conflict is a mistake, or that every new product is a gain for the people who bear its costs.

The pie is his analogy, and on the tape he half-apologizes for overusing it. The table at the top of this world is not a pie, because a pie still looks like one object being sliced. A copied method is a different event. A seat is a third. Keep his question. Do not let the metaphor settle the cases.

In the same answer he says the most likely outcome is an age of abundance, and that this would not be only universal basic income but universal high income. A voice in the room meets the medical sentence with “Not if they haven’t got a job.”

He says jobs are going to change, that jobs have always changed, and that being a computer used to be a job: buildings full of people doing calculations, including bank interest, now done by digital computers. He says he does not think anyone is pining to go back to that. He calls the most likely outcome incredibly beneficial, a future you would want if you could look into it.

This page does not hand him that forecast. The 2021 tape asked whether the pie is fixed. This tape asks what happens if a thing people experience as scarce, medical care better than anyone in that room, including him, could be had by everyone in the world. That is a different claim from “the pie has grown.” It is still a claim about what might be created. It is not a description of care anyone has received.

Some scarcity is a problem of allocation. There is one excellent surgeon. Who gets her time? Some scarcity can be attacked by creation. Knowledge, tools, robotics, infrastructure, and institutions might multiply a capability that used to sit in one pair of hands. Whether they do is not settled by the wish that they might.

He is using “super intelligence” as the name of the meeting. The same day, President Trump signed an order that the executive branch use “Super Intelligence” and “SI” in place of “Artificial Intelligence” and “AI.” The two words in the answer follow that political name. They are separated, in a later part, from Nick Bostrom’s 2014 definition. The syllables do not merge the claims.

Earlier that day he quote-posted a short vertical of the meeting and wrote only those two words. That post was not the source of the sentence above. Smoothed texts from clip accounts were not used either.

What if the pie didn’t have to stay the same size? Keep the question. It is still not a proof, and a pie is still the wrong picture for a method that can be copied. The harder question comes after the institutions, not before them. If we can create abundance, what makes it a collective good?

Stephen Wolfram’s writings give one disciplined picture of many histories kept at once. A multiway system, in the project glossary, is a system in which rules define multiple possible successors for states. A multiway graph connects states to the successors you get by applying those rules. In a 2020 essay he notes that branches can merge: two paths of update can arrive at the same state.

In the 2021 essay on the ruliad he writes that rules applied in all possible ways often generate new states, “leading to branching in the graph,” and that there can also be merging when different states become the same one.

He describes observers as computationally bounded. The amount of computation an observer does is, in his words, “tiny compared to the computation going on in the world around us.”

Branchial space, as he sketches it, lays out states on different branches so that histories sharing an immediate ancestor sit near each other. The ruliad is his name for “the entangled limit of everything that is computationally possible.” He presents that as a definition while the mathematics is still being built, not as a finished proof.

The project’s technical introduction speaks of models with the potential to represent fundamental physics. This page does not promote that potential into settled physics.

Nothing here claims that a human intention selects a quantum branch, or that ethics is a rule for picking one. The picture is useful only as a language: many possible histories, some of which rejoin, an embedded looker who does not see the whole graph.

Possibility means not ruled out. Contingency means neither impossible nor inevitable. A prediction asserts that one still-open outcome will happen. Probability grades outcomes. It does not merely list them. Choice and agency need an agent. Path dependence means the route can matter later, even if some routes eventually meet.

When a new capability enlarges what can be done, the graph of available actions gets larger. What becomes lived history is not decided by the size of the graph. Institutions, material limits, prior choices, and what people actually attempt still do that work.

Plato’s Republic puts the question in a ring. Gyges finds a ring that makes him invisible. Glaucon’s challenge is that a person who could do wrong without being seen would do it. In Paul Shorey’s translation, that person would take what he wished even from the market, “enter into houses and lie with whom he pleased, and slay and loose from bonds whomsoever he would,” and conduct himself “as the equal of a god.”

The same passage says that every man, when he supposes himself to have the power to do wrong, does wrong. That is a claim inside an argument, not a finding from an experiment. The enforcement has been removed. What remains is the person.

New instruments do not automatically bring a new grasp of the people holding them. In a 1957 interview printed in C. G. Jung Speaking, Jung says that “the world hangs by a thin thread, and that thread is the psyche of man.” He goes on: “There is no such thing in nature as an H-bomb—that is all man’s doing. We are the great danger. The psyche is the great danger.”

In the 1959 BBC Face to Face interview he says, separately, that “the only real danger that exists is man himself.” These are interview sentences, not a treatise, and not the shorter lines that circulate on graphics. The hinge they support is narrow. A larger power in human hands is not yet a larger understanding of human hands.

The same gap sits under the tools of this decade. Systems now write, see, plan, and act across more of ordinary work than they did a few years ago. That is a description of present tools. It is not a claim that a particular threshold has been crossed.

Nick Bostrom, in 2014, defined a superintelligence, tentatively, as “any intellect that greatly exceeds the cognitive performance of humans in virtually all domains of interest.” That is a technical use. It does not say the capability is already here.

A different use of nearby words arrived as political speech. A White House release of 22 September 2026, summarizing remarks at the United Nations, speaks of rejecting any attempt to construct a “globalist scheme of control” for the artificial intelligence then being discussed, “hereinafter officially called ‘Super Intelligence.’”

That sentence renames a technology in government language. It is not Bostrom’s definition, and it is not evidence about what the systems can do. On 29 September the signed order directs the executive branch to use those words in place of artificial intelligence and AI.

Reports in the days between the United Nations remarks and that order disagree in small ways about documents and timing. This world does not referee that dispute. The order is the text that was opened. The pool answer quoted in the hinge uses the meeting’s name while forecasting abundant care. That forecast is not evidence about the systems either. A political name and a technical definition can share syllables and still not share a claim.

What the tools do next is not settled by what we call them. Capability does not decide what a civilization values. If the background model is that someone must lose, more capability is more capacity to take. If some situations are not that game, more capability is also more capacity to create, including institutions that keep taking from being the only move. Neither sentence is a forecast. Both describe what a model permits.

A sibling walk, The Meaning of Life, asks this same question inside a path a visitor can take. It treats what you do when no one is scoring you as a signal, and it keeps a score of its own. This chapter does not.

A score kept by a page is another witness. The question was what remains when that witness is gone. The door opens on the plate where he put the unobserved act beside a world that might be grading goodness before release. That plate is his hypothesis. It is not a test this world will grade, and it is not evidence for the ring, the psyche, or the definition above.

The sentence is small, and it has the same structure as the table. If another person’s flourishing enters your experience as gain, the relation is not the chip game. Nothing was removed from you in order for their well-being to count as yours. That does not make jealousy impossible, or sacrifice unnecessary, or conflict fake. It says the accounting of a life can include a term the zero-sum form has no place for.

A civilization, a commons, a friendship, a person watching their own fear of being diminished: four scales, one question. The posts do not prove the feeling. They name it. Whether it is true in a given hour is not something a stranger’s page can settle.