Anthropic’s record $81B IPO puts its safety pledge under Wall Street scrutiny
Anthropic’s record $81.34 billion IPO brings Wall Street scrutiny to its safety pledge, raising questions about computing costs, control, and disclosure.

Anthropic began trading on the Nasdaq under ANT on Nov. 19 after pricing an $81.34 billion IPO, the largest on record, as The Rundown reported. Shares in the company behind Claude closed their first session at $97.17, more than 17% above the $83 offering price, for a valuation of about $1.75 trillion, according to AP.
The offering included 980 million shares at $83. Anthropic’s final prospectus put its proceeds at $78.84 billion after underwriting discounts and before other expenses. Settlement was expected on or about Nov. 20.
Before the opening bell, Anthropic’s CEO pledged to pace development around safety, saying the company “will not bend to either business nor political pressure.” That promise now comes with the scrutiny of public investors.
Why it matters
Two months ago, Sam Altman ruled out an OpenAI IPO in 2026, citing safety concerns and calling it an “ill-advised moment”. Anthropic went ahead. Its earnings calls will become a recurring test of both the AI boom and the company’s pledge to slow down if safety demands it.
The financial pressure has concrete terms. In its prospectus, Anthropic discloses minimum commitments of about $121.3 billion to Google through July 2033 and $110 billion to Amazon through April 2036, regardless of usage. Delaying a release could postpone revenue while those commitments create ongoing costs. Investors can ask how much room the company’s spending plans leave for a safety delay.
Control also shapes that test. The prospectus gives the Long-Term Benefit Trust authority to elect a board majority, while founders hold 50.1% of general shareholder voting power. Those arrangements limit ordinary shareholders’ control over the people making decisions about growth and safety.
Nasdaq describes Anthropic as a public benefit corporation, required to balance stockholders’ financial interests, affected stakeholders’ interests, and its public benefit mission. That gives its leaders a formal basis for weighing safety alongside returns. The test is how they exercise that authority when a delay carries a financial cost.
The safety pledge also needs a clear standard. Anthropic’s Responsible Scaling Policy, effective July 8, is voluntary and can change. It describes its safety roadmap as goals and distinguishes its own plans from more ambitious recommendations for the industry. Its prospectus says risk assessments inform its development and release schedules. Judging the promise will require watching how Anthropic applies that framework to specific decisions.
Customers planning products or services around future Claude models could face delays in their own work if Anthropic pauses a release. Clear explanations of timing and changes to safeguards would help them plan. They have a practical stake in understanding what could cause a delay and how Anthropic would communicate it.
Safety observers and investors also need a record of those decisions. The policy calls for Risk Reports every three to six months and external review in defined circumstances. Disclosures about delayed releases, changed safeguards, reviewer disagreements, and revised goals would help them judge whether practice matches the pledge. Reading that record alongside financial results would make earnings coverage more useful.
How management would resolve a future conflict between commercial targets and safety remains uncertain. Each earnings call offers a chance to ask what evidence would justify a pause, who could authorize a restart, and what customers and shareholders would be told.
Sources & further reading
- 01therundown.ai ↗
- 02A look at Anthropic's record-breaking IPO, by the numbers | AP News ↗
- 03www.sec.gov ↗
- 04At Nasdaq opening bell, Anthropic CEO recommits to pacing AI development | TechCrunch ↗
- 05Sam Altman: OpenAI won't go public this year as IPO now would come at an 'ill-advised moment' | Fortune ↗
- 06nasdaq.com ↗
- 07RSP v3.4 clean - Google Docs ↗
This story builds on reporting from The Rundown newsletter on November 20, 2026.