Anthropic's CEO asked the industry to slow down. The company's reported $100 billion IPO remains on an October schedule

Dario Amodei is calling for AI labs to slow the pace of capability gains while Anthropic pursues what could be the largest IPO ever attempted.

In a September 12 essay, Anthropic's chief executive wrote that "we must slow the pace at which we improve the capabilities of AI models," warning that "a race to the bottom, spurred by commercial incentives, can make these risks more acute." Altman said OpenAI would match Anthropic's commitment to outside evaluators.

Reuters has reported that Anthropic is preparing an offering that could raise as much as $100 billion, with marketing expected to begin in mid-October and a listing targeted for the days before the November 3 midterm elections. Amodei's essay does not mention the IPO, investors, shareholders or Anthropic's valuation.

Sam Altman made the connection directly. The OpenAI chief executive told Fortune the same day that this was "an ill-advised moment to go public" because of safety concerns, and ruled out an OpenAI IPO in 2026.

What happened in the same week

The essay landed in the middle of the busiest ten days of Anthropic's listing process. On September 4, Reuters reported the timetable: a $15 billion revolving credit facility to close first, a public registration statement expected in late September, analyst meetings after that, marketing in mid-October at the earliest, and a listing completed before the midterms. Morgan Stanley, Goldman Sachs, JPMorgan and Citi are among the banks. Reuters said the timing could still move.

On September 11, Reuters reported that Nvidia (NVDA) was in talks to anchor the offering with as much as $10 billion. Anthropic is seeking to raise as much as $100 billion at a valuation near $2 trillion, more than double the $965 billion it set in its $65 billion Series H on May 28. Anthropic declined to comment and Nvidia did not respond.

On September 12, Amodei published the essay. On September 13, Business Insider reported, in a Reuters item, that Anthropic had chosen Nasdaq for the listing, citing one person familiar with the plans. Neither company has confirmed it. The same evening, Reuters relayed a Financial Times report that Anthropic had told shareholders its adjusted operating income would be positive for a second straight quarter, with gross margins above 80% before revenue shared with distribution partners such as Amazon (AMZN) and before the cost of training models. Reuters couldn't verify the report, and the second quarter of profit is a forecast to shareholders, not a reported result.

What Amodei asked for, and what he did not say

The essay's one unilateral commitment is to embed third-party evaluators inside Anthropic with desks, badges, laptops and the right to publish findings "without editorial control by Anthropic." The rest depends on other companies and on governments: shared safety standards among labs in democratic countries, which he writes would need the US government to mediate or grant an antitrust waiver, and then a four-level arrangement with China. Pacing doesn't mean halting training, he writes. It means time spent on monitoring, alignment, interpretability and evaluation. The full proposal and his own list of reasons it will be hard are in the essay.

He gives no figure for how much slower. The essay says interpretability could make "profound progress in 1-2 years," and that a misaligned agent swarm could within "6-12 months" be capable of taking over the internet. It doesn't say what a slower Anthropic would spend less on, or whether a company raising $100 billion to buy computing capacity would buy less of it. Anthropic's compute agreements could reach $517 billion over about a decade, The Information estimated on September 6, a figure the company hasn't confirmed.

The company has addressed the question once, in a different document. In a policy post titled "When AI builds itself," Anthropic staff Marina Favaro and Jack Clark wrote that the company would slow down if other frontier developers did so "in a verifiable manner." It is conditional, and Anthropic has not said that it considers the condition met. Altman wrote on X that he agrees the industry must pace the frontier and that OpenAI will match the evaluator commitment, but he named no evaluator and no date.

The people who have connected safety and the listing

Altman is the most prominent executive to connect safety and listing timing on the record. "I'm happy to delay OpenAI's IPO in order to do what's required for safety," he told Fortune. Asked whether 2026 was off the table, he said, "I would say not 2026. Yeah, we got a lot of stuff to do, like meeting this moment of what is going to be required for safety and alignment." OpenAI filed its own confidential S-1 in June, and its chief financial officer told employees in August the company would be public in 2027.

Three days before the essay, Jacob Coxon, a pretraining researcher, resigned from Anthropic and wrote on X that "neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives." He left two months before his Anthropic stock vested, he told Axios, adding, "I no longer have anything to gain by juicing up Anthropic's valuation." Evan Hubinger, who leads alignment science at Anthropic, replied on X that Coxon was correct.

Anthropic hasn't commented on the timetable, the exchange, the anchor talks or the profit forecast. A public registration statement, expected in late September, may provide the first detailed account of Anthropic's safety practices, capital needs and risk factors in a document filed under securities law. Where the listing stands on any given day is on our AI IPO tracker. For a sense of what earlier holders paid, one Korean venture firm bought Anthropic shares at an $18 billion valuation.