Protect the Irreversible: What Matters Most in an Uncertain Future? | How to Think About the Future Pt 6, Frankly 151 artwork

Protect the Irreversible: What Matters Most in an Uncertain Future? | How to Think About the Future Pt 6, Frankly 151

The Great Simplification with Nate Hagens

July 17, 2026

This week's video is the sixth part in Nate's ongoing series, How to Think About the Future. Today, Nate turns from describing possible futures to exploring how we can make decisions when certainty is impossible.
Speakers: Nate Hagens
**Nate Hagens** (0:00)
Good morning. This is part six of How to Think About the Future, the final one, probably. So quick recap of where we've been. In part one, we talked about how to hold the future as a landscape of possibilities, coupled systems and phase shifts and the like. In part two, we built four grids, economic growth or contraction, power and distribution, geopolitical interconnection and earth system integrity. In part three, we explored the shape of the landscape itself with valleys and ridges and the asymmetry of transitions and how switchbacks and erosion make futures more or less likely. In part four, I highlighted four composite worlds, the long repair, Mordor persists or more and more Mordor. Forks of Swirled and the Unraveling. And I made the case that the future's most worth building will look like failure to a culture calibrated to growth. And that confusion between healing and dying at the cultural level may be one of the deepest challenges we face. Part five was a bit of an aside, applying these same civilizational themes to us as individuals on our own paths to the future. Throughout all this, I can imagine a question is building. Yeah, yeah, come on, Nate. What do I do given all this? Well, I'm not going to tell you that today. What I'm going to talk about is how to think about what to do. The logic of planning under uncertainty, which was my original ambitious intent with this entire now quite long series. I'm not going to highlight a list of specific actions or causes to support. Those specifics matter, and I have spoken about them elsewhere and will continue developing that work on this platform. But I think what will be most helpful to you all, whether you agree with the composites I've put forward or not, is a methodology for finding your own plan of action based on what you care about, your local context, what you want for the future. Because without such a plan, even well-intentioned actions can be unhelpful or counterproductive, and that's actually where I want to start.
Before we talk about how to prioritize what to do, I'd like to talk about how I think about what not to do. We can fail in two directions, and the causes of those failures look very different from each other. There's a concept from statistics, type one and type two errors, that mathematically describe positives and negatives.
Basically, imagine you hear a noise in the middle of the night, and you assume it's a burglar. Obviously, you were worried for your safety, so you called the police, but it turns out it was just your cat knocking stuff over in a kitchen. This is a type one error, a positive. You overreacted to a threat that wasn't real, creating an entirely different problem. Now imagine the flip side. You hear a noise in the night, and you assume it's a cat, because your cat, especially my cat, Toofy, is always running around knocking stuff over in the night, and you roll over and go back to sleep. Only this time, there actually was a burglar, and you missed it because you overlooked the warning signs, exposing yourself to a dangerous situation. This is a type two error, a negative. You missed a real threat because you didn't act. Both are dangerous. Both have consequences. They both happen all the time. And the challenge of us planning for a hazy and uncertain future involves both of these risks. Type two errors happen due to omission, things we miss. They are the shortfalls that we create by failing to act. Not protecting topsoil or biodiversity because GDP doesn't measure it. Not building community resilience because there's no monetary or institutional incentive to do that.
Not transitioning our energy systems because the current ones still technically work.
These are the things we should have done and haven't, at least not yet. But by the time the shortfalls become visible, it's often too late to reverse course. In the landscape language from part three, these are the washouts from cultural erosion. They're the switchbacks that we let disappear through neglect and inaction. On the other hand, Type 1 errors are the shortfalls we create by doing the wrong thing, acting on a model of the world that is incorrect or incomplete. These are positives. False positives come in a few flavors. The obvious one is acting on a partial map of our situation, like shutting down fossil fuels overnight, when 80% of the world still runs on them and 99.5% of our supply chains. Diagnosis is partially correct, but the prescription kills the patient. Then there's fighting the wrong war entirely, pouring resources into things like carbon offsets that don't offset anything, or ESG scores that don't change the behavior of the economic superorganism. This is all action, but action that's aimed at the wrong target. Said differently, the burglar is real, but we called the fire department instead of the police. And then there's an even deeper one, the deepest one actually. Nearly every institution on earth, every government budget and every pension fund and institutional infrastructure plan and corporate strategy and all the international agreements are built on the assumption that the economy of the future will be larger than it is today.

15 more minutes of transcript below

Feed this to your agent

Try it now — copy, paste, done:

curl -H "x-api-key: pt_demo" \
  https://spoken.md/transcripts/1000651996090

Works with Claude, ChatGPT, Cursor, and any agent that makes HTTP calls.

From $0.10 per transcript. No subscription. Credits never expire.

Using your own key:

curl -H "x-api-key: YOUR_KEY" \
  https://spoken.md/transcripts/1000777206894