A HARDENED CASE FOR HOPE

If you followed this all the way through, you may have noticed something strange.

The problem is far more complicated than it first appears—and the solution may be much smaller.

Once you strip away the epistemics, coordination theory, finance, bureaucracy, geopolitics, propaganda, regulatory capture, institutional self-preservation and all the other machinery, a huge portion of the problem reduces to something almost embarrassingly primitive:

A relatively small number of people and institutions are extracting enormous value from systems they have enough power to keep difficult to correct.

Beneath them are millions of perfectly ordinary people protecting smaller pieces of the same machine because their job, status, pension, contract, constituency, company, political coalition or mortgage depends on it. Nobody has to be running the whole thing.

Everyone only has to defend their piece.

That gives us a much cleaner split in the solution.

Either some of the most powerful beneficiaries realize that taking a little less can make the entire system vastly richer, safer and more durable—including for them—or we build enough distributed corrective power that nobody gets to permanently veto an obvious correction simply because they own the chokepoint.

Not a stronger dictator. Not our team capturing the machine.

Enough protected people, evidence, money, legitimacy, technical capacity and coordination that telling the truth and fixing something stops being a career-ending act of heroism.

At the highest level, the strategy is painfully simple:

Figure out where we're fucking ourselves. Figure out why we can't stop. Then make stopping possible.

And there is an almost incomprehensible amount of room to work with.

The United States is carrying roughly $40 trillion in public debt. In 2025, we spent $970 billion just servicing it—more than the United States spent on national defense.

Think about that for a second.

The interest on our debt now costs more than the largest military apparatus on Earth.

And even determining where enormous amounts of that money went has historically been a problem. In fiscal year 1999, the Pentagon recorded $2.3 trillion in unsupported or improper accounting entries. Another $1.8 trillion wasn't even reviewed.

That wasn't the end of it, but it was certainly one of the most spectacular examples.

Over the decades that followed, trillions more would accumulate in improper payments across the federal government—and the Government Accountability Office says the true amount may be significantly higher.

Then there is actual fraud.

GAO's best government-wide estimate puts direct federal losses to fraud at roughly $233 billion to $521 billion every year.

And that's before we get to waste, cost overruns, regulatory capture, legal extraction, unnecessary complexity, defensive spending, or the enormous amount of human intelligence spent fixing problems that never should have existed in the first place.

Extraction survives because nobody experiences the whole extraction stack at once.

Everyone can defend their individual five percent.

Nobody has to defend the combined forty, fifty, sixty percent.

Central architects, deliberate coordination, and captured institutions certainly exist. But they are not required for the larger pattern to emerge. Every node can optimize its own extraction while complexity hides the aggregate cost.

And every additional layer creates another constituency whose livelihood now depends on the complexity remaining in place.

Eventually, complexity stops being a problem to solve once enough people make their living from it. Now we have an incentive problem on top of the complexity problem—with whipped cream and a cherry on top of that.

Then the whole system gets $40 trillion worth of fucked, with no signs of stopping.

Eventually, the bill comes due. Maybe through inflation. Maybe austerity. Maybe financial crisis. Maybe the currency itself eventually breaks.

But at this scale, when the numbers finally stop being numbers, people get hurt. People lose homes, savings, businesses, futures—and sometimes much worse.

The deeper problem is not that nobody knows how to fix any of this.

A great deal of the intelligence already exists. So do many of the solutions. But the people holding different pieces are fragmented, professionally constrained, politically sorted, financially dependent, afraid of rocking the boat—or simply too jaded to believe trying will matter.

Speaking plainly can get you fired, canceled, isolated, sued, defunded or quietly removed from the room. Politics increasingly rewards defending your coalition over asking three basic questions: What is true? What actually works? What is the best solution we can build from here?

Because the bad menu itself is often the product of accumulated failure. We’re standing at the end of fifty bad decisions pretending the remaining two options are reality.

So enormous amounts of intelligence get spent defending bad systems instead of asking why these are the only choices on the menu.

Course-correct the underlying system and suddenly the question changes.

We are no longer arguing over which failure we prefer. We get to start building solutions that are intelligent, innovative, generative—and increasingly win/win.

The intelligence exists. The solutions exist in pieces. The people exist.

What is missing is permission, protection and convergence.

Which is why the first meaningful move can be almost comically small.

Not four people who are supposed to fix civilization. Four unusually capable people in one protected room where the normal incentives are temporarily suspended—where they can say what they actually think, disagree without punishment, connect domains that normally never talk, separate legitimate function from accumulated bullshit, and begin building a shared, updatable map of the terrain.

Protection matters here in more ways than one. People are much more willing to climb down into dark, dangerous holes and do important work when they have a good map, a light, and a solid group of people above ground holding the rope.

The first task isn't to fix ten million actors. It's to create one protected place where a few unusually capable people can finally look at the whole machine without having to pretend the machine isn't there.

In a strange way, the First Room is simply a permission structure for competence.

At sufficient altitude, the solution really does get painfully obvious.

Which is why I’d like to formally invite everyone to:

There are, of course, approximately ten million details between four capable people sitting in a room and peace in the Middle East... but that’s exactly my point.

Recognizing the destination is not the same as understanding the road.

A child can understand that people should stop doing things that make everyone poorer, less free, less safe, and eventually dead. The hard part has never been thinking of “stop.”

It is understanding what has to be true for stopping to become strategically possible while allowing people to remain about as good—and as terrible—as they naturally tend to be.

Fear, incentives, rivalry, information failure, concentrated power and coordination traps can make continuing destructive behavior perfectly rational from inside the game—even when the collective outcome is insane.

The answer is not to make humanity wiser everywhere.

It is to become strategically less wrong in the few places where error compounds—and strategically a little more right in the places where innovation, abundance and human capability become self-reinforcing.

Those dynamics were already expensive when our tools were primitive.

Add nuclear weapons, synthetic biology, autonomous systems, ubiquitous surveillance and increasingly capable AI, and the margin for stupidity starts disappearing.

Daniel Schmachtenberger compressed the problem beautifully:

“Rivalrous game dynamics, multiplied by exponential technology, are self-terminating.”

“Exponential tech is inexorable. We cannot put it away. So we either figure out anti-rivalry at scale, or the human experiment comes to completion.”

The inverse matters just as much.

The first country—or coalition—that becomes meaningfully less stupid about this doesn't need to achieve world peace. It may only need to reduce the worst 20–30% of its own extraction, waste, defensive bullshit and zero-sum behavior. That releases capital, talent, trust and productive capacity; those gains compound; and suddenly the society wasting less of its intelligence fighting itself has more intelligence available to build everything else—including, perhaps critically, the first general superintelligence that isn't born inside a game teaching it to dominate its competitors.

We are talking about technologies capable of producing almost unimaginable power and abundance.

And the admission price may be astonishingly low:

Stop using quite so much of that power against ourselves.

(For the alternative strategy, see The End of the World*.)

And the funny part is that most of the actual fixes probably aren't conceptually brilliant.

Stop paying ten times what something should cost. Stop creating artificial scarcity where abundance already exists. Stop allowing obvious conflicts of interest. Stop designing rules that reward extraction over production. Stop spending billions repairing damage caused by another thing we're spending billions to create.

Let people expose failures without getting destroyed. Let competent people cross institutional and ideological lines without having to perform loyalty rituals.

None of this is quantum mechanics or rocket science.

The correction itself can be painfully obvious. The hard part is everything that has accumulated around it: incentives, dependencies, contracts, careers, constituencies, power, and millions of perfectly reasonable reasons for the person standing at each individual tollbooth to explain why their tollbooth needs to remain exactly where it is.

The hard part is not always finding the correction. It's making an obvious correction possible.

And we do not need to fix all of it.

We don't need to end war. We don't need to eliminate greed. We don't need billionaires to stop being billionaires. We don't need nations to suddenly trust one another, politicians to stop being politicians, or eight billion people to achieve enlightenment somewhere on a fucking mountain.

We need to become meaningfully less stupid in a few places that matter enormously.

Even five percent matters. Because that five percent doesn't just sit there.

It is the point where the interest starts compounding in a thousand different ways across the length and breadth of the system. Money that isn't wasted becomes capital. Time that isn't consumed becomes creation. Better information improves thousands of decisions downstream. Less defensive burden frees people and resources to build. Greater abundance reduces some of the pressure driving zero-sum behavior in the first place.

And now each of those changes begins interacting with the others.

The five percent isn't the payoff. It's where the compounding starts.

Everybody understands compound interest when the thing compounding is money.

We seem to forget that human capability compounds too.

Which brings the whole thing down to a surprisingly small first step.

Stop starting over.

Give the people who already hold different pieces of the answer a shared map that can survive disagreement—one that preserves what has actually been established, exposes what is still genuinely disputed, kills the same bullshit escape hatches without having to kill them again six months later, and gets better every time reality teaches us something new.

Then let competent people do their jobs.

Not from zero. Not through another wall of tribal arguments, rhetorical games and institutional memory loss. From the best map we have so far, with enough protection and coordination to act on what it shows.

THE WAY IS THERE.

After more than twenty years of looking at how people and systems go wrong, I am not particularly easy to sell hope to anymore.

And yet, after everything we just looked at, after all the damage already done, after getting this close to the edge—I still think the door is open.

A genuine golden age of prosperity and cooperation is still possible—even if our current course is taking us somewhere else.

There is, however, one final trap buried underneath almost everything else in these pages.

If enough people become convinced meaningful correction is impossible, they stop paying the costs required to attempt it. Then impossibility becomes partially self-fulfilling.

People don't risk careers for something they believe cannot work. They don't invest money. They don't cross ideological lines. They don't endure ridicule. They don't build institutions. They don't cooperate with people they distrust.

They retreat into whatever smaller game still seems survivable.

Which is why the first meaningful corrections matter so much.

The first step is not to save civilization.

It is to stop losing the ground we've already won.

And that may be the final thing worth understanding.

The mechanism is not morally directional.

The same feedback dynamics that can carry a civilization downward can carry it the other way.

Perhaps for the first time in human history, we may actually have the knowledge, technology, productive capacity and connective intelligence required to make a radically better civilization possible at scale.

Not perfect.

Not free of greed, stupidity, conflict, corruption—or people being people.

Better. Significantly better in all the ways that really matter, not just materially.

A world in which more people are free to build lives they actually want, and the systems around them are better able to see what is true, correct what is wrong, and survive the parts of human nature that aren't going anywhere.

For most of my life, I have been looking for some version of the same three things:

A way out of some of what we've built.
A way back to some of what we've lost.
A way through what comes next.

For the first time in my life, I believe fundamental systemic change is actually possible.

That does not mean we get there.

It means the path exists—perhaps in a way it never has before in human history.

That distinction changes everything.

The same human forces can produce the horrors of this world—or something far better than what we have learned to expect.

Both are possible. Both are human.

And after all of this, what we finally have… is a choice.