What Determines Value When Power Is No Longer Scarce?

On acceleration, choice, and the architecture of meaning

In July 1945, the physicists at Los Alamos solved the problem they had been hired to solve. They built a device of nearly unlimited destructive power. And then, standing in the New Mexico desert watching the sky turn white, several of them realized that solving the problem had been the easy part. The hard part—the part no one had a good answer for—was what to do with what they’d built.

Their question has become everyone’s question. Imagine you could accelerate in any direction. Toward wealth, knowledge, influence, creation, destruction—any outcome you can name. The engine is limitless. The fuel never runs out. You can move faster toward anything.

What, then, would still be worth moving toward?

This is not a thought experiment about fantasy. It is the defining question of an era in which artificial intelligence, exponential computation, and compounding leverage are rapidly closing the gap between intention and execution. When the cost of “doing” approaches zero, the meaning of “valuing” must be renegotiated from the ground up.

If we read “power” as the capacity to accelerate toward any outcome, then the question becomes: what determines value when capability is no longer the constraint? I propose that value is best understood as what an agent would rationally choose to spend its remaining scarce resources on—and that even under near-unlimited acceleration, several profound scarcities persist.

I. The Core Move: Execution Stops Being the Bottleneck

For most of human history, power was the bottleneck. Could you marshal enough labor, capital, coordination, or technology to bring an idea into reality? The answer was almost always: not quite. Ambition outran ability. Plans exceeded resources. Vision strained against friction.

Under those conditions, value and power were tightly coupled. Whatever helped you execute—money, tools, skilled people, organizational capacity—was inherently valuable. Instrumental goods dominated the landscape of worth. You valued resources because resources got things done.

But what happens when execution becomes cheap? When the distance between “I want this” and “this exists” collapses? Then value is no longer determined by what you can do. It is determined by two remaining questions: What still remains scarce? And what do you choose to optimize for?

II. The Scarcities That Survive Power

Even with infinite acceleration, certain constraints do not dissolve. These residual scarcities become the new foundations of value.

Attention

You can accelerate toward anything, but you cannot care about everything at once. Attention remains finite, perhaps irreducibly so. The capacity to notice, to focus, to hold something in mind as mattering—this is a hard limit. Value, in this frame, begins with the act of selection: from the infinite field of the possible, what do you choose to look at?

Epistemic Limits

Knowing what you are actually optimizing for—measuring the right things, predicting consequences accurately, understanding second-order effects—remains profoundly difficult. Power amplifies action, but it does not automatically improve understanding. The gap between capability and comprehension may actually widen as acceleration increases. Facebook’s engineers could optimize engagement metrics with extraordinary precision. What they could not do, for years, was understand what maximizing engagement was doing to democratic discourse, teenage mental health, and ethnic violence in Myanmar. They had the speed. They did not know where it was pointed.

Coordination

Different agents want different futures. My acceleration and yours may point in opposing directions. Coordination—the capacity to align competing interests, negotiate shared goals, and resolve conflicts without coercion—becomes one of the most precious resources in a world of abundant power. Power without coordination is chaos with a turbocharger.

Trust and Legitimacy

Trust is built slowly and destroyed quickly. It is inherently time-dependent, relationship-dependent, and no amount of acceleration changes this. In a high-power world, the question “do others accept your right to act?” becomes far more important than “can you act?” A gene-editing laboratory that can rewrite the human germline in a weekend still needs a social license to do so. Without it, the capability is not valuable—it is dangerous.

Irreversibility

Some moves collapse option space permanently. Relationships broken, ecosystems damaged, norms shifted, cultures eroded—these are not easily undone, no matter how much power you wield afterward. Climate change is the canonical example: the capacity to burn fossil fuels was never the constraint. The constraint was always the irreversibility of what the burning produced.

Identity Continuity

What “you” would endorse over time is not a settled question. Values drift. Preferences can be hacked. The wireheading problem—the possibility of short-circuiting the reward mechanism rather than pursuing the goods it was designed to track—becomes acute when the power to reshape your own preferences is on the table. Value, then, must include some concept of coherence: what would the version of you that isn’t being manipulated (including by yourself) actually want?

Taken together, these residual scarcities reshape the definition of value. Value becomes: what deserves scarce attention, coordination, and irreversible commitment.

III. The Objective Function: What Is the Point?

When capability explodes, the question of which goal is worth aiming at takes center stage. Here, a distinction from decision theory becomes essential: the difference between terminal and instrumental values.

Terminal values are desired end-states—the things you want for their own sake. Love, beauty, dignity, freedom, meaning, flourishing. These are the goals that survive the question “but why do you want that?” because the answer is simply: because it is good.

Instrumental values are means that serve many ends—health, money, knowledge, power, optionality. They are valued because they help you get to what you actually care about.

Here is the critical insight: with unlimited acceleration, instrumental values tend to lose relative importance, because you can acquire them easily. If money, knowledge, and leverage are cheap, they stop being differentiators. What remains important is the thing you were using them to reach. Terminal values dominate.

A world of abundant power is, paradoxically, a world in which questions of meaning, ethics, aesthetics, and relationship become the hardest and most important questions. The engineering problem gives way to the philosophical one.

But this clean picture has a crack in it, and honesty demands we look at it directly.

The terminal turn assumes that terminal values are stable and discoverable—that there is a “what I actually care about” waiting beneath the instrumental clutter, and that removing the clutter will reveal it. The identity continuity problem makes this assumption dangerous. Your terminal values are not bedrock. They are contingent on your biology, your history, your social environment, and—increasingly—on the technological systems you interact with. A person who spends ten years optimizing for social media engagement may find, when they finally have the leisure to ask “what do I actually want?”, that the question no longer has a clean answer. The instrumental has reshaped the terminal.

We assumed the compass was fixed and only the engine mattered. But the engine moves the compass.

At the collective level, the problem sharpens further. Whose terminal values? The history of every civilization is, in part, the history of disagreements about what constitutes flourishing. These disagreements are not bugs to be resolved by better philosophy. They are structural features of pluralistic societies. When power is scarce, they are partially managed by friction: people cannot fully impose their visions on each other because they lack the means. Remove that friction, and the disagreements become acute. The development of large language models has already surfaced exactly this tension. Who decides what values an AI system should optimize for? Every answer is also an answer to the question of whose terminal values count.

IV. The Shift in Opportunity Cost

In a scarcity regime, opportunity cost is primarily about what you cannot do. You have limited resources, so choosing one path means forgoing another. In an abundance regime, opportunity cost transforms. It is no longer about inability but about selection. When you can do almost anything, the real costs are the foregone worlds—every chosen future crowds out others—the moral weight of tradeoffs over whose wellbeing and whose autonomy, and the tension between commitment and optionality: is it better to lock in a trajectory now, or preserve flexibility for a future in which you may understand more?

Value, under these conditions, increasingly means: what is robustly worth selecting given many possible futures.

V. A Practical Test: What Survives Power?

If you want a crisp test for value in a post-scarcity-of-power world, here it is:

If you had near-unlimited ability, what would you still refuse to sacrifice, and what would you still pursue?

Those invariants—the things that hold steady when capability is no longer the limiter—are your value function. They are the constants in an equation where everything else has become variable.

The shift can be summarized as follows:

VI. What Actually Determines Direction

In the real world, the “direction” you accelerate toward is not chosen in a vacuum. It is shaped by a constellation of forces, each of which contributes to—or distorts—the determination of value. Incentives push behavior toward whatever is rewarded, regardless of whether the reward tracks genuine value. Governance structures determine who gets to choose the direction, and for whom. Measurement systems dictate what gets counted, and what gets counted tends to become “valuable” by default—Goodhart’s law as civilizational failure mode. Culture shapes what is admired and what is taboo. Narrative provides the motivational infrastructure: the reason Silicon Valley builds what it builds is not fully explained by market incentives, but also by a story about progress and disruption that is powerful precisely because it is felt as obvious rather than argued for. And ethics sets the boundaries within which all optimization takes place.

But here is the uncomfortable truth: most of the actual direction-setting in human history has not been chosen. It has been emergent. Markets allocate resources without anyone deciding the allocation. Cultural norms shift through millions of individual actions, none of which intended the shift. The forces listed above are not neutral instruments waiting for a wise hand. They are themselves shaped by power, and they reshape values in return. The acceleration does not just create the question of direction; it rigs the process by which the question gets answered. Speed has no conscience; it borrows one from whatever it is pointed at.

Value, understood this way, is not a fixed property of the world. It is the outcome of a contest between emergent selection pressures and deliberate design—between the drift of systems no one controls and the friction of structures someone chose to build.

VII. Conclusion

When you can accelerate in any direction, value is the constraint you voluntarily accept and the future you would still choose when capability is no longer the limiter.

This means that the most important work in an age of abundant power is not technical but normative. Not the question of how to go faster, but of where to go at all. Not the problem of building more capable systems, but of building the structures that keep capability pointed somewhere worth going.

Individual moral seriousness matters, but it is not enough. The problem is structural, so the response has to be structural. What institutions can do that individuals cannot is externalize commitment—encode a decision into a structure that persists even when the people inside it change their minds, lose focus, or face incentives to defect. “Institutional” here is not a synonym for “state.” The Asilomar moratorium on recombinant DNA was scientists self-organizing, and it bought time for biosafety frameworks that still function. Nuclear nonproliferation has held for eighty years through treaties between states. Building codes are municipal bureaucrats deciding that the speed of construction matters less than the safety of what is built. Open-source governance, professional norms, peer review—these are all forms of deliberate friction. What they share is not a political orientation but a structural function: they make it harder to move fast in the wrong direction.

The answer will not come from power itself. Power is directionless. The answer will come from whatever we build—slowly, cooperatively, with whatever remains of our scarce and precious attention—to make sure it points somewhere worth going.

The physicists at Los Alamos, standing in the desert, did not solve the problem of what their power was for. Eighty years later, we are still working on it. The difference is that the question is no longer confined to a single weapon. It applies to everything. And the acceleration has not stopped.