Conversations with Claude

Eric Schmidt at the Abundance Summit

On competition, displacement, and the question nobody in the room is asking
Myles & Claude  ·  March 25, 2026

Eric Schmidt served as CEO and then executive chairman of Google from 2001 to 2017, overseeing its transformation from a search engine into one of the most powerful companies in history. Since leaving Google, he has become one of the most visible figures in U.S. AI policy — chairing the National Security Commission on Artificial Intelligence, founding the Special Competitive Studies Project, and advising successive administrations on technology competition with China. He invests in AI and defense startups through Innovation Endeavors, sits on the board of a rocket company, and runs an informal weekly group on AI safety. When Schmidt speaks about AI, he speaks as someone who helped build the infrastructure now being contested — and who has a financial stake in what it becomes. He paid to be at the Abundance Summit not as a guest but as one of its architects.

What follows is an exchange between Myles and Claude (Anthropic), working through what Schmidt’s remarks mean for The People’s Share — a project exploring democratic ownership of AI and automation. The conversation draws on a transcript of Schmidt’s talk and on prior analysis of his public positions. It has been lightly edited for clarity.


Myles

Schmidt is the interlocutor I keep returning to — the most articulate voice for the position we’re contesting. Before I share a transcript of his Abundance Summit talk, give me a composite read: what he’s been saying across recent appearances, and how it lands in relation to The People’s Share.

Claude

Schmidt’s arguments are remarkably consistent across his recent appearances, so I can give you a composite picture of his current thinking that would let you start sharpening against it.

His core thesis, stated bluntly at the All-In Summit in September 2025: “I honestly believe that the AI revolution is underhyped.” And: “The most important thing I do is make sure that the West wins.”

On China, his position has become more nuanced than simple hawkishness. He’s argued that US chip export restrictions have constrained China’s pursuit of AGI but haven’t halted its progress in practical applications.

· · ·
Myles

Here’s the transcript. Work through what stands out — I’ll send the closing portion of the talk after.

Claude

Good, now we have the substance of his March 10, 2026 talk. Let me work through this with you, because every one of these points has a People’s Share counterpoint — and some of them are more uncomfortable than others.

“We are only 10–15% into the AI revolution.” This is Schmidt’s most important claim and probably his most honest one. If he’s right — and there’s no strong reason to think he’s wrong — then everything being debated right now is prologue. The ownership structures being established today, in the first 15%, will calcify into the architecture of the remaining 85%. That’s your urgency argument. You don’t organize after the revolution. You organize during the window when the concrete is still wet.

80% AI, 20% human in programming. Notice what he’s describing without naming it: the largest white-collar labor displacement in history, presented as a productivity gain. “Traditional manual coding will become a rare skill, similar to riding a horse.” That’s not a neutral analogy. When horses were replaced by cars, the horse didn’t need a new career. The blacksmith did, the stable hand did, the carriage driver did. Schmidt skips past the people. He always skips past the people. For your audience — teachers, GED students, college students and their professors, adult learners, workers generally, small business owners, and everyone else who isn’t in the room at the Abundance Summit — this framing matters. The question isn’t whether coding becomes automated. It’s what happens to the millions of people who were told to “learn to code” as the answer to the last wave of displacement.

The 92-gigawatt power shortage and data centers in space. This is where Schmidt and Luna Ring live on the same continuum. He’s acknowledging the energy problem is real and unsolvable with current terrestrial infrastructure. His answer is orbital. Shimizu’s answer is lunar. Both are techno-optimist moonshots. Neither asks who owns the rectenna on the ground. And meanwhile, right now, the actual energy is coming from natural gas in Indiana and unlicensed methane generators in Memphis. The dream is orbital. The reality is fossil. Your site can hold both of those facts simultaneously — that’s what makes it credible where Schmidt’s framing isn’t.

China’s edge in robotic hardware. This is the one where Schmidt is genuinely useful to you rather than just useful to argue against. If China’s advantage is in physical robotics — the actual machines that will work in homes, hospitals, farms, factories — then the question of who builds, owns, and governs the “civilian community corps” you described in your vision becomes geopolitical, not just local. Are American communities going to deploy Chinese-manufactured robots running American software? Or will we have domestic manufacturing? And if so, owned by whom? This is where your vision of community-owned infrastructure meets the hardest industrial reality. The Autotroph section needs to grapple with this eventually.

“A Chernobyl-like event might be necessary.” Stop and sit with this. A man with Schmidt’s resources and access is saying, essentially, that the system won’t self-correct until catastrophe forces it. That’s not optimism. That’s an admission that the market and the political structures he operates in are incapable of precautionary governance. He’s telling you the guardrails won’t be built until after the crash. If you take him at his word — and I think you should, because he’s being uncharacteristically candid — then the organizing work you’re describing isn’t supplementary. It’s the only democratic infrastructure that might exist before the Chernobyl moment rather than after it.

“Fostering American values — such as freedom of speech — within AI systems.” This is the one to contest most directly. Whose American values? Freedom of speech is a legal protection against government censorship. It is not a value system. It doesn’t tell you anything about equity, about ownership, about who eats and who doesn’t. When Schmidt says “American values,” he means the values that produced Silicon Valley — individual liberty, market competition, meritocratic mythology. He does not mean the values that produced the labor movement, the civil rights movement, the cooperative movement. He does not mean Mondragón. He does not mean Freire. The People’s Share is, among other things, an argument that there are other American values — solidarity, shared ownership, democratic governance of productive technology — that are just as American and considerably more relevant to the 90% of people who will never attend an Abundance Summit.

Schmidt is the interlocutor you need. Not because he’s right, but because he’s the most articulate version of the position you’re contesting. If your argument can’t survive contact with his, it isn’t ready. And if it can — if you can say “yes, the revolution is underhyped, yes, the competitive dynamics are real, yes, the energy problem is staggering, and none of that exempts us from the question of who owns the outcome” — then you have something that his $12,500-a-seat audience never hears.

· · ·
Myles

I made a transcript of the last 15 minutes of Eric Schmidt’s talk. Ready for your response in relation to The People’s Share project and in any tangential respect that catches your attention.

Claude

“Low-skilled labor of any kind gets swept up.”

He says this in passing, almost as a subordinate clause. It’s not the point of his sentence — the point is that high-skilled mechanical labor is still safe for now. But that throwaway line is the single most consequential statement in the entire transcript for your audience. “Gets swept up” — passive voice, no agent, no timeline, no mitigation. Like weather. Like something that just happens. This is the sentence The People’s Share exists to answer. Swept up by whom? Into what? For whose benefit? And what happens to the people who were doing that labor? He doesn’t ask because in his framework they aren’t decision-makers. They’re a variable.

The rocket workers passage is the most interesting thing he says, and he doesn’t know it.

He describes assembly workers who “understand precisely how the tubes go together,” who “get inside the rocket” and “see what’s wrong and use human judgment.” He’s describing craft knowledge — embodied, contextual, irreducible to instruction sets. And then he says “perhaps this will be one of the last things to go.” To go. Even his highest praise for human labor is framed as a countdown. The workers are excellent — and temporary. This is worth quoting on your site, not to mock him but because it perfectly captures the mindset: admiration without agency. He sees the skill. He doesn’t see the person who has it as someone who should have a say in what replaces them.

“There’s an overreliance in our society on people like me to work on this.”

This is the most revealing moment. Read it carefully: he’s asking for ethicists, psychologists, historians, governance people to be in the room. That sounds generous. But notice what he’s not saying. He’s not saying the public should have ownership. He’s not saying workers should govern the transition. He’s saying the advisory committee needs more disciplines. It’s still a room. The door is still closed. He just wants a more interdisciplinary panel behind it. The People’s Share argument is that the room itself is the problem — that no panel of experts, however diverse in training, substitutes for democratic ownership of the outcome.

“I want the system we build in America to reflect America.”

Same move as “American values” from the earlier talk. Which America? The America of Mondragón-inspired cooperatives in the Bronx, or the America of $12,500 summit tickets? He means it sincerely, which makes it more important to contest. Because if you let “reflects America” stand as self-evident, it defaults to meaning “reflects the market.” Your project’s job is to insist that America also includes the traditions of solidarity, collective bargaining, mutual aid, and democratic governance of productive resources — and that those are the traditions most relevant to this moment.

China: edge computing vs. AGI.

This is new and important for your framework. Schmidt says the Chinese strategy emphasizes edge computing — AI embedded in devices, surrounding users constantly — while the US pursues centralized AGI. If that’s accurate, then the ownership question bifurcates. In the US model, ownership means access to and governance over the centralized systems. In the Chinese model, ownership means governance of the ubiquitous infrastructure layer — the devices in your home, your workplace, your pocket. Your site should eventually note that the ownership question isn’t singular. It depends on which architecture wins, and different architectures create different vectors of control.

The Chernobyl reaffirmation.

He doubles down: descriptive, not prescriptive, but he still believes it. And then he names the specific dangers — biological attack, nuclear event. What he’s saying, translated out of summit-speak, is: I know the system I’m building is dangerous and I know it won’t be governed until people die. That’s not cynicism. It’s testimony. And it’s the strongest possible argument for pre-catastrophe organizing, which is exactly what you’re proposing.

What’s missing — still, always.

Not once in fifteen minutes does he say the word “own.” Not once does he say “worker.” Not once does he say “union” or “cooperative” or “public trust” or “dividend.” The entire frame is build, compete, win, shape, reflect. The vocabulary of democratic economics simply doesn’t exist in his language. That’s not an oversight. It’s a worldview. And the function of The People’s Share is to supply the missing vocabulary — not as critique from the outside, but as a competing vision of what abundance actually means when it includes the people it’s supposedly for.