Summary

Thiel’s base case is that AI will be internet-sized, not a total transformation of society. He estimates the internet added a few percentage points to GDP and perhaps 1% annual growth for 10–15 years; AI may do something comparable, creating great companies without ending stagnation. The investable case rests on AI being “the only thing we have,” not on a guaranteed superintelligence cascade.
The bottleneck to progress may be institutions and heterodoxy rather than raw intelligence. Society already has many smart people yet remains stuck, leading Thiel to doubt that higher AI “IQ Elo scores” automatically produce dementia cures, factories, or Mars colonies. A conformist AI could instead generate infinite acceptable content, displace workers, and deepen stagnation—but he still prefers that risk to “nothing at all.”
The wider economy remains stagnant because progress after 1970 concentrated in bits while atoms, medicine, transport, and energy slowed. Thiel contrasts 1750–1970’s accelerating ships, trains, cars, aircraft, Concorde, and Apollo with 1985–2025’s superficially similar built environment. His practical tests are whether 30-year-olds live better than their parents did and whether research yields justify rising inputs.
Growth is not merely upside in Thiel’s model; it is what keeps middle-class expectations, fiscal bargains, and political institutions coherent. He defines the middle class as people who expect their children to do better, while describing Reagan’s “consumer capitalism” and Obama’s “low-tax socialism” as growth-dependent contradictions. Without renewed dynamism, taxes rise or socialism ends; a static “Greta future” would be a radically different and oppressive settlement.
Trump remains “by far the best option” chiefly because disruption can clear the field for construction—not because Thiel expects effortless technological greatness. Silicon Valley’s shift followed years of increasing the liberal “dose” until companies spent hundreds of millions, “go completely woke, and everybody hates you.” Nuclear deregulation offers a possible path from deconstruction to building, but Thiel now calls political involvement both “incredibly important” and “incredibly toxic.”
AI undermines the fantasy that entrepreneurs can escape politics by moving offshore or to Mars. Elon Musk told Thiel, “There’s nowhere to go,” while Demis Hassabis’s warning that superhuman AI could follow Musk to Mars underscored that Mars might not function as an alternative political society. The same technology could deter China from Taiwan by widening its disadvantage—or accelerate an invasion under a “now or never” calculation.
Thiel’s ultimate tail risk is centralized rule sold as protection from technological catastrophe. His formulation, “Antichrist or Armageddon,” links nuclear, climate, bioweapon, and AI fears to global compute control capable of logging every keystroke; the FDA’s and Nuclear Regulatory Commission’s de facto worldwide influence are his milder precedents. Douthat’s strongest challenge is that such totalitarianism may first require the very AI, surveillance, and military systems he says Thiel is investing in.

Deep dive

1. The stagnation thesis survives AI

Thiel says he still “broadly” believes the stagnation thesis, which never meant zero progress. His claim is about velocity: from 1750 through 1970, change accelerated across ships, railroads, cars, and planes, culminating in Concorde and Apollo; afterward, “in all sorts of dimensions, things had slowed.”
Bits remain the major exception—computers, software, the internet, mobile internet, crypto, and now AI. AI is “in some sense, pretty big,” but Thiel leaves open whether it can overcome generalized stagnation rather than merely create another narrow vertical of progress.
Measuring stagnation is intrinsically difficult because late-modern knowledge is hyper-specialized. An outsider cannot easily score string theory against quantum computing or cancer research, while ever-narrower groups of expert “guardians guarding themselves” make independent judgment harder; for Thiel, that opacity is itself grounds for skepticism.
Douthat’s common-sense test is the time traveler: someone moving from 1890 to 1970 would encounter another world, while children in 2025 see 1985 as recognizable except for phones and modestly different cars. Thiel’s economic test is equally concrete: compare a 30-year-old’s living standard with that of their parents at 30.

2. Growth holds the middle class and its institutions together

Thiel does not dismiss the fears that helped halt acceleration. Continued progress plausibly appeared to point toward environmental or nuclear apocalypse, while institutions also became risk-averse and people ran out of ideas; his account therefore combines legitimate danger, cultural change, and institutional decay.
He defines the middle class as people expecting their children to do better than they did. Once that expectation collapses, the middle-class society collapses with it; perhaps a static feudal order could function, but not the United States as it operated through its first 200 years.
Fiscal institutions likewise assume growth. Thiel calls Reaganism “consumer capitalism”—capitalists borrowing instead of saving—and Obama’s model “low-tax socialism”; he prefers the latter to high-tax socialism but says it cannot remain stable indefinitely. “At some point, the taxes go up or the socialism ends.”
Douthat asks whether genuine degrowth could provide a stable alternative. Thiel allows that a “Greta future” might work, but it would be a radically different society and “super oppressive”—not necessarily North Korea, yet far removed from the comfortable status quo its advocates may imagine.

3. Escaping decadence requires more risk, especially in medicine

Douthat’s warning is that anti-decadence can become hunger for crisis: after September 11, some foreign-policy conservatives imagined war could awaken and remake a complacent country, with disastrous results. Thiel answers directionally rather than precisely: “We should take a lot more risk. We should be doing a lot more.”
His sharpest example is dementia and Alzheimer’s research, where he says there has been “zero progress in 40 to 50 years.” Researchers remain stuck on beta amyloids despite failure, creating what he calls “some kind of a stupid racket” sustained by institutional self-reinforcement.
More risk could mean giving patients with fatal diseases and their researchers wider latitude, but Thiel’s deeper appeal is cultural. Early modern thinkers expected disease cures, radical life extension, and even immortality; science had to rival Christianity’s promise of physical resurrection rather than retreat into marginal improvements.
At PayPal around 1999–2000, co-founder Luke Nosek’s interest in Alcor led the company to a “freezing party” selling cryonics policies for full bodies or cheaper heads. The malfunctioning dot-matrix printer supplied the stagnation joke, but Thiel’s point is serious: fringe boomers still believed science might defeat death, while Thiel says no millennial believes that anymore.

4. Trump was a wager on diagnosis before delivery

Douthat frames Thiel’s political turn as venture investing in disruptive agents. Thiel accepts the analogy but distinguishes ambitions: perhaps Trump could redirect “the Titanic” away from the iceberg; more modestly, “Make America great again” could force acknowledgment that America was no longer great.
Thiel says he had few positive expectations for Trump in 2016, but welcomed the first Republican in a century who rejected “this syrupy Bush nonsense.” He now calls his hope for a real conversation a “preposterous fantasy,” recalling two incompatible intuitions: nobody would care if Trump lost, yet Trump had a 50–50 chance of winning.
What changed was less policy than diagnosis. In debates with Eric Schmidt in 2012, Marc Andreessen in 2013, and Jeff Bezos in 2014, Thiel argued stagnation while each offered a version of “everything’s going great”; he says all three, and Silicon Valley more broadly, later updated to varying degrees.
Thiel describes the liberal default’s escalation: when it failed, companies increased the dose, spent hundreds of millions, went “completely woke,” and became hated. The eventual pivot was not necessarily pro-Trump; it was recognition that the prior approach was not working.

5. Populism can clear the field, but construction remains unproven

Asked whether Trump 2.0 can deliver technological dynamism, Thiel says it is “still by far the best option we have.” Douthat’s pushback—worth keeping—is that this sounds like an argument that disruption cannot be worse, not evidence that populists care about science or will fund it.
Thiel replies that funding unreformed institutions is not synonymous with supporting science. A modern Einstein’s letter would disappear in the White House mailroom, while today’s “moonshot” means “something completely fictional that’s never gonna happen”—the reverse of the 1960s, when a moonshot meant reaching the moon.
His hoped-for sequence joins deconstruction to construction: deregulate nuclear power, then eventually build new plants, better designs, or fusion reactors. Clearing the field is necessary but insufficient; the political project must attempt both, and Thiel concedes the two stages are deeply entangled.
That entanglement helps explain his retreat from direct political funding. He is “schizophrenic on this stuff”: politics is simultaneously “incredibly important” and “incredibly toxic,” trapping everyone involved in a zero-sum world whose stakes feel exceptionally high.

6. Mars no longer provides an exit from politics

In 2024, Thiel told Musk that a Trump loss might make him leave the country. Musk answered, “There’s nowhere to go. This is the only place”—prompting Thiel’s delayed realization that Musk had stopped believing in Mars as a political escape, even if he might still regard it as a scientific and technical project.
Thiel traces the change to an earlier exchange he helped broker with Demis Hassabis. Hassabis called superhuman AI the world’s most important project; Musk countered with making humanity interplanetary. Hassabis replied that his AI “will be able to follow you to Mars,” and Musk went quiet.
In Thiel’s telling, Musk took years to process the implication: a “woke AI” or socialist government could follow settlers off-world. Once Mars is concretized as an alternative society rather than merely a rocket program, technological escape no longer solves the underlying political problem.

7. AI may be internet-sized without unlocking the physical world

Thiel’s “stupid answer” is that AI is “more than a nothing burger” but less than a total societal transformation. His placeholder is the late-1990s internet: perhaps a few percentage points added to GDP and roughly 1% extra annual growth for 10–15 years—enough for major companies, perhaps not enough to end stagnation.
Douthat tests the superintelligence cascade: could smarter models cure Alzheimer’s, automate factories, build a billion robots overnight, and unlock other forms of progress? Thiel doubts intelligence is the gating factor because society already contains many smart people who perform worse as intelligence rises, cannot apply it, or do not fit institutions.
He offers two darker explanations: dementia and mortality might be genuinely unsolvable, or cultural systems might suppress the heterodox people needed for “crazy experiments.” If AI is merely conventionally smart and conformist, its intelligence may not make a difference.
Douthat’s feared outcome is an AI that makes infinite okay movies and ideas, eliminates jobs, yet deepens stagnation. Thiel agrees that is “certainly a risk,” but invokes the internet: it did not produce the cornucopian, diverse explosion libertarians fantasized about in 1999, yet was probably still better than the counterfactual. Without AI, “there’s just nothing going on.”

8. Silicon Valley debates capability while neglecting meaning

In the exchange, Douthat hesitates when asked whether humanity should survive; Thiel presses the question and answers yes, while also insisting on radical solutions to human problems.
The transhumanism discussion treats transformation of clothes, sex organs, or the body as insufficiently ambitious; Thiel adds that Orthodox Christianity demands transformation of the soul too. Douthat counters that Christianity promises perfected body and soul through grace, whereas builders of a machine god generally do not see themselves cooperating with God.
Thiel replies that Judeo-Christian inspiration is about transcending fallen nature. Douthat says uploading is a step down from cryonics: “I’d rather have my body” than a program simulating him.
The mismatch between rhetoric and belief appears in Musk’s prediction of 1 billion humanoid robots in the United States within 10 years. Thiel told him that such growth would erase budget-deficit worries; Musk remained worried, suggesting he had not fully connected the claims, expected less economic transformation, or recognized “big error bars.”
Silicon Valley’s weakness, Thiel argues, is interpreting technology’s meaning. It obsesses over AI “IQ Elo scores” and exact AGI definitions while neglecting budgets, growth, and geopolitics. AI could deter China from invading Taiwan because China has effectively fallen behind—or accelerate invasion because Beijing concludes it must act now.

9. “Peace and safety” could turn existential risk into global control

Thiel reframes nuclear war, climate disaster, bioweapons, and AI as risks that invite another “bad singularity”: a one-world totalitarian state. Global compute governance might control every computer and “log every single keystroke” to prevent dangerous AI, raising the question whether protection moves society “from the frying pan into the fire.”
His paired formulations are “one world or none” and “Antichrist or Armageddon.” The Antichrist’s implausible path to power becomes plausible if a leader talks incessantly about Armageddon, then promises regulation and safety; in today’s fearful culture, Thiel says the likelier figure is not Dr. Strangelove but Greta Thunberg.
Douthat presses the central contradiction: universal safety rule may require an actual technological shock and the AI, military, surveillance, and Palantir systems capable of enforcing it. Thiel rejects the suggestion that this is his project, but argues that 50 years of worldwide “peace-and-safety-ism” already offer a moderate version of the pattern.
His evidence is regulatory reach: other countries effectively defer to the FDA on drugs and the Nuclear Regulatory Commission on reactors, while Thiel says nuclear power has been off-ramped worldwide. Stagnation may reflect exhausted ideas, bottom-up docility, or top-down machinery that made experimentation impermissible.

10. Human freedom, not providential inevitability, closes the argument

Douthat asks whether this means society already lives under a moderate Antichrist and whether God will eventually end the stagnation. Thiel resists determinism: history leaves “room for human freedom and human choice,” and attributing every event to God risks “scapegoating God.”
Citing Christ’s “They hated me without cause,” Thiel argues against making divine causation the explanation for persecution or political failure. Douthat pushes back that God’s entry into history through Christ is basic Christianity, not Calvinism, and asks for hope that stagnation will not be permanent.
Thiel’s practical conclusion is action rather than fatalism. If the outcome were predetermined, one might practice “yoga and prayerful meditation” while waiting for the lions to eat them; that is not what people are supposed to do. He and Douthat finally agree that resistance should carry hope of success.

Daniel Gross and Nat Friedman: Acquired by Meta | Microsoft Layoff 9000 People | OpenAI's Bombshell Michael Dell – Invest America Act Becomes Law, AI Talent Wars, Compute Demand, Market Update | BG2