FULL STORY

AI Giants Call to Slow Down, Markets Tumble

Top AI executives led by Dario Amodei called for slowing the frontier, sparking controversy, confirmed talks on an industry standards body, and a global AI stock selloff.

2026-09-13 ~ 2026-09-15 · 11 episodes · 86 posts

Episode 1 · Debate Rages Over Antitrust Waivers for AI Safety Collaboration (2026-09-13, 6 posts)

A fierce debate has broken out between U.S. policy circles and the AI industry over whether AI safety collaboration should receive antitrust exemptions. Anthropic CEO Dario Amodei publicly called for a narrow waiver so labs can discuss safety without antitrust exposure, after reports that OpenAI had asked members of Congress about the same issue. Policy researcher Will Rinehart filed comments with DOJ and the FTC proposing a limited safe harbor, citing last summer's joint safety evaluations by OpenAI and Anthropic. On the other side, former FTC commissioner Terrell Bedoya published a sharply critical essay.

Confirmed

  • Dario Amodei called on the government to issue a narrow waiver so AI labs can talk about safety without antitrust risk (m4)
  • OpenAI was previously reported to have asked members of Congress about antitrust issues around AI safety collaboration (m4)
  • Will Rinehart filed comments with DOJ/FTC proposing a limited antitrust safe harbor for AI safety, citing the OpenAI–Anthropic joint safety evaluations from last summer (m5)
  • Former FTC commissioner Bedoya argued that warnings about AI extinguishing humanity should not be taken seriously (humanity survived the ice age, the Black Death, two world wars and nuclear weapons), and that if a company truly believed its product had a 10% chance of killing people, the right response would be to stop work (m1, m3)
  • Bedoya said individual employees' panic may be sincere but misguided, and that CEOs seeking broad antitrust exemptions should be viewed with deep skepticism given the industry's economic landscape and competitive threats from open-source models (m2)

Unconfirmed

  • No regulator has responded on whether any waiver will materialize; former antitrust official Ioana's view is an argument under existing antitrust frameworks, not an official position (m6)

Why it matters

The fight exposes a core tension in AI governance: safety advocates say labs need legal channels to share risk information, while antitrust voices worry that "safety" could become a pretext for top companies to dodge competition oversight. Bedoya's paradox cuts deep—if a company truly believed in a fatal risk, why seek a collaboration waiver instead of halting work. Ioana's suggested path (competitor agreements that increase output can be permitted, and "preventing catastrophic risk" can be argued as increasing output, as DOJ previously argued for cybersecurity collaboration) offers legal room for a compromise, showing the issue remains openly contested.

Episode 2 · AI Slowdown Signals from Top Lab CEOs Spark Bubble Fears (2026-09-13, 4 posts)

After Musk, Sam Altman and Dario Amodei signaled AI slowdown, commentators predicted a 10-20% drop in AI stocks on Monday, calling it the AI bubble bursting, while some buyers offered to scoop up shares at a discount.

Episode 3 · Anthropic CEO Dario Amodei Says He'd Hand the Company to the Right Combination of Governments (2026-09-13, 9 posts)

Anthropic CEO Dario Amodei publicly said he would be willing to put advanced AI, and some form of control over the company, into the hands of "the right combination of governments," adding that it has always felt "weird" and uncomfortable for a private company to build AI. The rare statement is seen as a landmark signal in the relationship between frontier AI labs and governments, and has drawn debate over its governance logic.

Confirmed

  • Asked on CBS Sunday Morning whether he would hand control of the company's AI to government, Amodei said he wasn't sure it would be "handing over control," but some form of oversight or joint governance—taking years to build—could be the path.
  • Per a Forbes interview, he would give control of AI to "the right combination of democratically elected governments."
  • He said it feels uncomfortable that AI is built by a private company, consistent with his long-standing pro-regulation stance, amid global jockeying over frontier AI control.

Unconfirmed

  • Some posts suggest this fuels speculation about a possible US government stake in Anthropic; BecauseCulture joked it might mean Anthropic's valuation (in its regulatory-filing sense) is heading down. Neither claim is supported by evidence in the materials.

Why it matters

  • A frontier lab CEO voluntarily offering to cede some control to governments is nearly unprecedented, touching the core dispute over frontier-model safety and state collaboration.
  • Not everyone is convinced: Rob LeClerc sarcastically questioned why governments—holding a monopoly on violence and lacking market-based checks—would be more fit than private firms to steward AI.
  • AlexTensor went further: if international governance means the public backstops Anthropic's commercial risks, a 50% equity tax is far too little—the public should hold a controlling stake, and outright nationalization must be on the table. The spat shows "who controls frontier AI" remains the industry's deepest fault line.

Episode 4 · Dario's 'Hand Over Anthropic' Remark Taken Out of Context: He Meant Joint Oversight (2026-09-14, 2 posts)

Polymarket reported that Dario Amodei would hand Anthropic to 'the right combination of governments,' sparking debate. Critics noted the quote was taken out of context—his original words pointed to some form of joint government oversight rather than a full handover.

Episode 5 · OpenAI, Anthropic and Google Reportedly in Talks to Form Industry-Led AI Standards Body (2026-09-14, 10 posts)

According to The Information reporter Leo Schwartz, OpenAI, Anthropic, and Google—the three leading frontier labs—have been in regular discussions about establishing an AI industry standards body, with the most recent meeting held last week and talks still ongoing. This is the first time top AI companies have been reported to jointly push a self-regulatory framework for the industry.

Confirmed

  • The source is a report by The Information, with multiple reposts pointing to the same origin; @Hesamation added that the three parties have held regular discussions since July on setting up an industry-wide standards organization
  • The discussions center on creating an industry-led standards body that would share model safety protocols and enable self-regulation of the AI industry
  • The report draws an analogy to FINRA (Financial Industry Regulatory Authority), the self-regulatory model used in the financial industry
  • The coordination began before Dario Amodei's public post (in which he called for slowing the pace of development)

Why it matters

  • If it materializes, this would be a major step for AI safety governance, with leading companies jointly setting standards that could profoundly shape future industry safety frameworks and the regulatory landscape
  • One reposter (@StewartalsopIII) sees the move as a gesture of collaboration amid competition; others raise concerns about the boundary between industry self-regulation and public regulation

Episode 6 · Dario's Call to Slow Frontier AI Sparks Global Governance Debate (2026-09-14, 39 posts)

Around September 14, Anthropic CEO Dario Amodei published "We Must Pace the Frontier," calling for a slowdown of frontier AI. It quickly triggered a chain reaction across the industry and from US and Chinese officials, sparking a rare global debate over AI governance approaches with sharply divided positions.

Confirmed

  • Sam Altman, Elon Musk, Demis Hassabis, Satya Nadella and others publicly responded within hours of Dario's post (as compiled by The AI Daily Brief podcast).
  • Altman published a piece the same day outlining two catastrophic AI paths: humans losing control of AI, and superintelligent AI power being monopolized by individuals or companies. He stressed that competitive pressure is no excuse for recklessness, that capabilities must not outrun alignment and oversight, and welcomed a unified federal frontier AI safety regulatory framework, supporting independent audits and saying the industry can self-regulate without waiting for legislation.
  • OpenAI Chief Scientist Jakub Pachocki said no lab has currently solved alignment and oversight well enough for long-term full-speed scaling, advocating a voluntary slowdown and shared safety standards.
  • Gary Marcus relayed The Information's report that AI companies had already been discussing Dario's proposed ideas behind the scenes; he himself questioned whether a company genuinely close to AGI would sincerely propose slowing down.
  • Cohere CEO Aidan Gomez published "Who Gets to Define the Rules for AI?," criticizing the idea of two or three Silicon Valley companies acting as a global government's AI rule-setters as a "cartel," and advocating evidence-based standards with broader participation.
  • According to CNBC, China responded to the slowdown calls from multiple AI company CEOs, calling them "fear marketing"; Reuters reported that Chinese state media (including Global Times) published pieces attacking Anthropic's slowdown call as a "Cold War tactic" / "silent AI Cold War," arguing that the side already in the lead is the one most comfortable calling for a slowdown.

Unconfirmed

  • The specific content and progress of companies' behind-the-scenes discussions come only from The Information's secondhand account; details have not been made public.
  • Critics' interpretations (such as blogger boriquagato and Airtable founder Bindu Reddy) of Dario's motives — using regulation to block open-source competition and protect profits — are personal speculation, and Dario's side has not responded.

Why it matters

  • This is the first time a top-lab CEO has explicitly called for slowing the frontier and received responses from multiple peers as well as OpenAI's own Chief Scientist, potentially signaling a shift toward industry-coordinated safety commitments.
  • The slowdown call drew criticism from both Chinese officials and the US conservative camp; as @vista8 noted, Dario "gets no buy-in from either side," highlighting the structural conflict between US-China AI competition logic and safety governance demands.
  • The open-source camp and companies like Cohere worry regulation could be weaponized by incumbent giants into a "cartel"; the core dispute is: who should set the rules, and should regulation target AI capabilities themselves or specific applications.

19 more related posts →

Episode 7 · AI Executives Call to Slow Development, Global AI Stocks Plunge (2026-09-14, 6 posts)

Leaders of several top AI companies have issued rare public warnings about the risks of AI moving too fast—seen as the most direct threat yet to hundreds of billions of dollars in AI investment—and AI-related stocks worldwide fell sharply in response. Current takeaway: industry leaders have shifted from breakneck expansion to collective warnings, and markets have reacted with real money; the fallout from this turn deserves close attention.

Confirmed

  • Anthropic CEO Dario Amodei published "We Must Pace the Frontier," urging the industry to slow "reckless" AI development
  • Per Reuters, Altman publicly backed Amodei's slowdown call, and Google DeepMind's Demis Hassabis and Elon Musk among others echoed it, citing safety concerns
  • Per Reuters, Asian AI-related stocks plunged Monday: SoftBank opened down as much as 13.2%, Kioxia fell 9.8%, SK Hynix dropped 5.3%; per the Financial Times, AI-concept stocks fell as much as 13% in a single day
  • Per The Guardian, investors are re-evaluating AI companies' valuations; AI stocks had previously driven global markets to repeated record highs
  • Per The Guardian, Altman said OpenAI will not go public this year

Why it matters

  • This is the first time top-lab executives have so collectively and publicly hit the brakes on their own industry, a stark reversal of the previous "accelerate and expand" narrative
  • AI stocks were a key driver of record highs in global markets; the sell-off suggests investor confidence in AI valuations and narratives may be wavering
  • Poster koltregaskes observed that the frontier labs' successive statements over a few days served as the direct catalyst for this shift in market sentiment

Episode 8 · Palisade Director Tells BBC the World Is Waking Up to AI Risks (2026-09-15, 2 posts)

Jeffrey Ladish, executive director of Palisade Research and former Anthropic safety engineer, told BBC World that "the world is starting to wake up" to AI risks, advocating embedded audits alongside national oversight in response to CEOs' slowdown calls.

Episode 9 · Debate Flares Over 'AI Is Slowing Down' Narrative (2026-09-15, 2 posts)

KeyTryer pushed back against claims that AI progress is slowing, predicting more capability shocks this year, while physicist Sabine Hossenfelder argued that frontier labs' sudden willingness to slow down stems not from existential risk but poor safety measures.

Episode 10 · AI Labs' 'Slow Down' Pleas Spark Backlash as PR Stunt (2026-09-15, 3 posts)

Multiple AI labs' coordinated calls to 'slow down AI for humanity' drew sharp criticism, with veteran commentator David Linthicum calling it the most cynical PR move in his 30-year career. Critics argue truly committed labs should simply stop racing and instead accelerate openly published safety research.

Episode 11 · OpenAI Confirms Weeks of AI Safety Talks with Anthropic and Google (2026-09-15, 3 posts)

OpenAI policy chief Chris Lehane confirmed the company has been coordinating with Anthropic and Google DeepMind on AI safety for weeks, a rare collaboration among frontier labs, though some argue the resulting slowdown could benefit Chinese rivals.