SteelmanPodcast

Launching soon;)

[Sean McClure @sean_a_mcclure]

https://t.co/1EzLtoBiuq

[SteelmanPodcast @SteelmanPodcast]

@Briankeating any podcast with you as host and @sean_a_mcclure as guest is indistinguishable from magic... It's critically important we have substantive conversations on AI. Sean is your guy. You're welcome.

[SteelmanPodcast @SteelmanPodcast]

@Briankeating any podcast with you as host and @sean_a_mcclure as guest is indistinguishable from magic... It's critically important we have substantive conversations on AI. Sean is your guy. You're welcome.

[Sean McClure @sean_a_mcclure]

❌Humans are not formally verifiable.

❌Companies are not formally verifiable.

❌Governments are not formally verifiable.

❌Markets are not formally verifiable.

Yet we manage them through incentives, monitoring, redundancy, oversight, constraints and institutional checks.

No one demands a mathematical proof that a CEO will never make a harmful decision before allowing companies to exist.

AI isn’t introducing “uncontrollable intelligence” into society for the first time. Society is already full of emergent optimization processes.

A corporation is already a kind of distributed intelligence. No single employee controls it. The corporation’s behavior emerges from thousands of interacting incentives, procedures, and local decisions.

Likewise, markets are already nonhuman optimization systems, bureaucracies are already nonhuman decision systems,
institutions already exhibit emergent behavior that nobody explicitly intended.

Error is tue cost of genuine complexity, and there is nothing more complex than systems that produce general intelligence.

The control theorists are treating intelligence as a program verification problem, when intelligence is actually a complexity-management problem.

Error is not a bug. Error is the unavoidable consequence of operating in a space that is too complex to be exhaustively searched, predicted, or verified.

Evolution produces errors constantly. Markets produced bubbles constantly. Science produces false theories constantly. Human cognition produces biases constantly.

You cannot navigate this space to perfect reasoning.

If you achieve the guarantees that you want, you would have a dumb system, by definition.

These podcasts keep talking to the wrong people; reductionists who have no understanding of complexity. You cannot talk about AI properly without understanding complexity.

[SteelmanPodcast @SteelmanPodcast]

Universal Generational Wealth (UGW) paid for by a Fully Endogenous Coasian TfP. For the record.

[Sean McClure @sean_a_mcclure]

If you take what AI just gives you, it's going to be disjointed, incomplete, somewhat misrepresenting and at times downright dumb.

But this is no different than walking into a meeting with experts and taking the first ideas that come to mind.

Like all good work, it requires a copious amount of dialogue to cancel out errors, and converge to something that is good, and at times even great.

AI is not an answer machine, and treating it as such is dumb. It is a dialogue machine, which can work with people to arrive at unique, solid and at time innovative work.

Pointing to AI's mistakes as evidence for lack of intelligence or genuine creativity is a straw man. The one thing all great human creators have in common is an absolute ton of dumb mistakes when zoomed in. Those errors are *critical* ingredients to true creativity. They produce the gaps, the juxtapositions. Access to the information one needs to create great work is not available without them.

Ai is not a "machine" in the sense dissenters talk about it. It is not rules-based, and it is not there for "answers." It is there for dialogue.

What survives that dialogue is what creativity is. It does not exist anywhere else. It is not something that came from AI, it is something that materialized from persistent conversation.

If you are using AI correctly, the outputs that are produced do not exist in the conversation. They stand outside it, induced by one's experiences and abilities.

Do not take what AI gives you. That's dumb. That is not how to use AI, just as that is not how to "use" humans.

It's already inside you. You just need to do a lot of talking to get it out.

[Sean McClure @sean_a_mcclure]

https://t.co/9fuAuVC6D2

[Sean McClure @sean_a_mcclure]

In AI, the intelligence isn’t in the optimizer.

It’s in the geometry of the loss landscape it moves through.

Latest lesson:

https://t.co/iZIR5EtMgC

[Sean McClure @sean_a_mcclure]

https://t.co/b1ddv9woAS

[SteelmanPodcast @SteelmanPodcast]

@RazzberryYams @OfSymbols @thebadstats Of course it does. Think it through.

[SteelmanPodcast @SteelmanPodcast]

@thebadstats @OfSymbols Abstractions are *necessarily* broad, so you could be forgiven for misreading Sowell's aphorism as commentary on P.O. but, if your dislike for irrelevant figures didn't prematurely terminate your own thoughts, you'd see its direct referent is Opportunity Cost - v. core to econ.

[Brett Hall @ToKTeacher]

🧵From “Bold Conjectures I” published by @ConjectureInst and edited by @ChipkinLogan :

@DavidDeutschOxf on “The four most fundamental theories”.

[Deutsch Explains @DeutschExplains]

.@DavidDeutschOxf: "Wealth is not a number. I don't think it can be characterized very well by a number. It is a set.

The set of all transformations that you are capable of bringing about, that is your wealth.

And obviously, if optimism is true, then there's no limit to wealth.

And at any one time, there is a rough correlation between the wealth that is the set of all transformations that you could bring about, and other things that aren't very fundamental, like the amount of money you have, or the amount of energy you control, or the amount of land you control, or the amount of power you have, and so on.

But those are not fundamental. They are all outgrown, eventually, by the growth of knowledge."

[SteelmanPodcast @SteelmanPodcast]

@harleyf In a strong field, @sean_a_mcclure's book, "Discovered, Not Designed" is, without question, the best and most impactful book you'll read this year. I'd be happy to send you a copy. I'm so sure you'll love it, I made a donation to @MUHCFoundation - I'll double it, if you read it.

[Santa Fe Institute @sfiscience]

New book: An Introduction to Ergodicity Economics

A new textbook by SFI External Professor Ole Peters and Alexander Adamou draws on physics to re-examine core assumptions in economic theory. It presents a formal challenge to expected-value thinking and shows how non-ergodicity can arise naturally in economic models.

The book introduces ergodicity economics as a “third way” alternative to expected-utility theory and behavioral critiques of rationality. It is the first textbook in this emerging field, and it offers a new way to think about decision-making under uncertainty.

https://t.co/YAsd5sonsk

[SteelmanPodcast @SteelmanPodcast]

Insane that one of the clearest thinkers today *only* has 10K followers. His great treatise, Discovered, Not Designed, was the book I gifted most in 2025. In the age of Social Media, with counterfeit talents flooding the exchange, his deep insights feel like alien gold. Congrats.

[Sean McClure @sean_a_mcclure]

A little milestone. Only took 15 years.

[Karl Popper Quotes @QuotePopper]

"Every solution of a problem raises new unsolved problems; the more so the deeper the original problem and the bolder its solution. The more we learn about the world, and the deeper our learning, the more conscious, specific, and articulate will be our knowledge of what we do not know—our knowledge of our ignorance. For this, indeed, is the main source of our ignorance: the fact that our knowledge can be only finite, while our ignorance must necessarily be infinite."

— Karl Popper, Conjectures and Refutations: The Growth of Scientific Knowledge (1963), p. 28.

[Ole Peters @ole_b_peters]

1/thread🧵
Almost 20 years ago, I started thinking about the ergodicity problem in the context of economics. That turned out to be surprisingly fruitful, and now there's a book about it.

[David Deutsch @DavidDeutschOxf]

Constructor Theory of Time: https://t.co/FgsECTHvTt

[Karl Popper Quotes @QuotePopper]

"The growth of knowledge depends entirely upon disagreement."

–– Karl Popper.

[Karl Popper Quotes @QuotePopper]

“True ignorance is not the absence of knowledge, but the refusal to acquire it.”

–– Karl Popper.