Key takeaways
- Anthropic priced its IPO at $83 a share on Nov. 18, 2026, selling 980 million shares for $81.3 billion, the largest IPO on record by base deal size. ANT stock was expected to begin trading on Nasdaq on Nov. 19.
- At $83, Reuters reported, Anthropic is worth about $1.1 trillion on shares outstanding and about $1.5 trillion fully diluted.
- The MisAnthropic Statement, published Nov. 18, says “Anthropic is not a safe AI company.” Its page listed 116 signers as of 04:11 UTC Nov. 19, including Steven Pinker, Max Tegmark and Guillaume Verdon (‘Beff Jezos’).
- Anthropic’s prospectus calls AI that helps build the next AI, which it terms “closing the loop,” “one of the most significant structural advantages in frontier intelligence,” while saying it does not expect to do that safely without meaningful human supervision.
- Anthropic’s seven co-founders hold 50.1% of the vote through a single Class F share, so public ANT buyers cannot outvote them.
Anthropic priced its IPO at $83 a share on Wednesday, Nov. 18, above its $70 to $78 range, selling 980 million shares, up from 825 million. The $81.3 billion raise is the largest IPO on record by base deal size in Renaissance Capital’s ranking, ahead of SpaceX’s $75 billion in June. ANT is expected to start trading Thursday on Nasdaq.
More than eight hours before Anthropic posted its price, a statement went up carrying the sentence no company wants on listing day: “Anthropic is not a safe AI company.” By 04:11 UTC Thursday its page listed 116 signers from camps that rarely agree, and Polymarket had pushed it to its 2 million followers. Anthropic’s prospectus calls the process the statement warns about a structural advantage.
What $83 a share buys
Anthropic’s IPO at $83 a share
| Item | Figure | Note |
|---|---|---|
| Price | $83.00 | Marketed range was $70 to $78 |
| Shares sold | 980,000,000 | Up from 825,000,000 in the Nov. 17 filing |
| Gross proceeds | $81.3 billion | 980 million shares at $83 |
| Sold by co-founders | 23,648,646 shares | About $1.96 billion; Anthropic gets none of it |
| Underwriters’ option | 147,000,000 shares | About $12.2 billion more if fully exercised |
| Private placement | $26.0 billion | Nonvoting Class N stock at the IPO price |
| Valuation | About $1.1 trillion | About $1.5 trillion fully diluted, per Reuters; $965 billion after the May funding round |
| Revenue, first nine months of 2026 | $33.2 billion | Up from $2.5 billion a year earlier |
| Loss from operations, same period | About $1.3 billion | Operating expenses were $34.5 billion |
| Run-rate revenue | About $81 billion | As of Oct. 31, up $10 billion from Sept. 30 |
| Committed spending | About $900.9 billion | $451.6 billion of it expected before the end of 2030 |
Proceeds, co-founder and option dollar figures and the operating loss are SnowRock’s arithmetic from the filing. Run-rate revenue is Anthropic’s annualized measure of recent activity, not reported revenue.
Source: Anthropic’s Nov. 18 pricing announcement; S-1/A No. 3, filed Nov. 17; Reuters.
At $83, Anthropic is worth about $1.1 trillion, or $1.5 trillion fully diluted, Reuters reported. On Sunday, before the deal grew, Bloomberg reported it was close to seven times oversubscribed. Watcher Guru, citing Bloomberg, had the price two hours before Anthropic.
Bloomberg Tech host Ed Ludlow’s Nov. 18, 2026 report, published before Anthropic’s announcement, that the IPO had priced at $83 a share and the offering had grown, citing people familiar with the matter.
Anthropic Is Said to Price IPO at $83 Per Share on YouTubeLargest IPOs by base deal size
| Company | Priced | Base proceeds |
|---|---|---|
| Anthropic | Nov. 2026 | $81.3 billion |
| SpaceX | June 2026 | $75.0 billion |
| SK hynix | July 2026 | $26.5 billion |
| Saudi Aramco | Dec. 2019 | $25.6 billion |
| Alibaba | Sept. 2014 | $21.8 billion |
| SoftBank Corp. | Dec. 2018 | $21.3 billion |
| NTT Mobile | Oct. 1998 | $18.1 billion |
| Visa | March 2008 | $17.9 billion |
Base proceeds exclude shares underwriters buy later. Counting those, SpaceX raised about $85.7 billion, Reuters reported; Anthropic would reach about $93.5 billion if its option is fully exercised.
Source: Renaissance Capital’s ranking of the largest global IPOs, retrieved Nov. 19, 2026; Reuters.
Who signed ‘not a safe AI company’
“Anthropic warns that recursively self-improving AI could concentrate power and escape human control, yet races to build it first,” the MisAnthropic Statement begins. It says the company “quietly weakened its safety pledges and lobbied for self-serving regulation” and that “we must reject its reckless attempt to amass power in the name of the greater good.” The Center for AI Safety, run by Dan Hendrycks, announced it; Hendrycks wrote that “numerous AI safety groups” agreed privately but felt unable to sign “without risking relationships or future funding.”
The names are the story: Steven Pinker; Hugging Face co-founder Thomas Wolf; xAI co-founder Igor Babuschkin; Guillaume Verdon, the accelerationist known as Beff Jezos; George Hotz; Lawrence Lessig; Nobel laureates Oliver Hart and Sheldon Glashow; Judea Pearl; and Max Tegmark.
Some went straight at the buyers. Psychologist Geoffrey Miller wrote that anyone “working for Anthropic, or investing in Anthropic” is “supporting one of the most dangerous and unethical companies in human history.”

Tegmark’s post, seen about 641,000 times by 04:11 UTC, says Anthropic’s leaders are “pedal to the metal on RSI” in private, “boasting to prospective investors” that it is their edge. It also says “WaPo reporting suggests that they’re implicated in more civilian deaths than any other AI company,” citing a school “selected for targeting by Claude as part of their Palantir collaboration.” SnowRock could not access that reporting and found no official finding that Claude chose the target.
Wolf said he “wouldn't have used ‘reckless’” but that a company “on track to be the most powerful company on earth” owes the public transparency “close to, or maybe even higher than, that of a sovereign government.”
The prospectus says it out loud
The filing calls “closing the loop,” in which “successive versions of an SI model assist in building the next,” “one of the most significant structural advantages in frontier intelligence.” SI is its shorthand for super intelligence.
Its risk factors warn that using AI to help build more capable AI “could also pose a considerable safety risk, particularly if it sharply accelerates the pace of progress,” and say “we do not anticipate being able to safely use Claude to fully autonomously design and build successive generations of frontier models without meaningful human supervision or understanding.”
David Sacks, who co-chairs the president’s science and technology council, wrote he had been “predicting an Anthropic-Doomer split,” saying Anthropic “has marketed ‘closing the loop’ on its IPO roadshow as a core advantage.” Signer Tristan Harris promoted a “Ban RSI rally outside Anthropic offices” in San Francisco for 11 a.m. Thursday. Signer Gary Marcus, naming no one, posed a choice: back “a messianic cult leader” or say no.
The backlash to the backlash
Researcher Nirit Weiss-Blatt, author of The Techlash, wrote the signers condemn Anthropic’s power grab and “don't see the irony,” since their own groups have sought “sweeping restrictions, expanded government enforcement, and even surveillance powers in the name of AI safety.”
Others cheered what the signers attack. Mechanize co-founder Matthew Barnett replied: “I stand with Anthropic, since they have done a lot to accelerate the race to broadly useful superintelligent AI agents while keeping safety a top priority.” Co-founder Ege Erdil endorsed it.
SnowRock found no public response from Anthropic as of 04:11 UTC Thursday. SnowRock has covered Anthropic’s October filing on catastrophic risk, Dario Amodei’s warning about AI you can’t switch off and an Anthropic researcher’s viral jobs clip.
ANT buyers get stock, not the steering wheel
A single Class F share, held by the seven co-founders’ LLC, carries “50.1% of the total voting power” on matters including electing the directors the trust does not pick. A Long-Term Benefit Trust elects directors who “together will constitute a majority” of the board. Amazon and Google, set to own about 14.7% and 13.7% before the deal grew, per the Nov. 17 filing, hold nonvoting shares.
The co-founders are selling about $280 million of stock apiece, and Dario Amodei will still beneficially own about 237 million shares, worth about $19.7 billion.

Seven chip, memory and cloud companies, including NVIDIA, Microsoft and Google, agreed to buy $26.0 billion of nonvoting stock at the IPO price. Anthropic has committed about $900.9 billion, “substantially all” of it to compute and infrastructure.
SnowRock’s read: the signers and the prospectus describe the same machine, AI that helps build the next AI. The signers call it reckless; the filing calls it an advantage it does not expect to pursue safely without meaningful human supervision. If they collide, seven co-founders and a trust decide, not ANT buyers.
Predictions: ANT’s first trade, Anthropic’s next policy
Prediction: ANT opens above $83 on Thursday, given that demand and a Hyperliquid pre-IPO contract, a bet on price, not shares, that traded near $100 late Wednesday, New York time. An open at or below $83 breaks this call.
Prediction: the signers will not slow the loop. By Feb. 19, 2027, Anthropic will update its safety policy “to keep humans in control,” as its filing says it is considering, without pausing AI that helps build AI. A pause, or no update by then, breaks this call.
Common questions
What price did Anthropic set for its IPO?
Anthropic priced its IPO at $83 a share on November 18, 2026, above the $70 to $78 range it had marketed. It sold 980 million Class A shares, up from 825 million in its November 17 filing.
When can you buy Anthropic stock, and what is the ticker?
Anthropic’s Class A shares were expected to begin trading on the Nasdaq Global Select Market on November 19, 2026, under the ticker ANT. Once listed, they can be bought through a brokerage account like other listed stocks. The offering is expected to close on November 20, 2026.
How much money did Anthropic raise in its IPO?
The IPO raised about $81.3 billion, 980 million shares at $83, including about $1.96 billion that goes to co-founders who sold shares. Separately, AMD, Broadcom, Google, Micron, Microsoft, NVIDIA and a TSMC affiliate agreed to buy $26.0 billion of nonvoting Class N stock at the IPO price in a private placement; those shares carry no underwriter lock-up but are restricted securities.
Is Anthropic’s IPO the largest ever?
By base deal size, yes: its $81.3 billion tops SpaceX’s $75 billion from June 2026 in Renaissance Capital’s ranking. Counting extra shares sold to underwriters, SpaceX raised about $85.7 billion; Anthropic would pass that if its underwriters buy roughly 53 million or more of their 147 million optional shares.
What is Anthropic’s market cap at its IPO price?
At $83 a share, Anthropic is worth about $1.1 trillion on shares outstanding and about $1.5 trillion including options and other dilution, Reuters reported. Its May 2026 funding round valued it at $965 billion.
What is the MisAnthropic Statement?
It is a short public statement published November 18, 2026, the day Anthropic priced its IPO, that says “Anthropic is not a safe AI company” and accuses it of racing to build recursively self-improving AI. The Center for AI Safety, run by Dan Hendrycks, announced it.
Who signed the MisAnthropic Statement?
As of 04:11 UTC on November 19, 2026, the statement’s page listed 116 signers, including Steven Pinker, Thomas Wolf, Igor Babuschkin, Guillaume Verdon (Beff Jezos), George Hotz, Max Tegmark, Gary Marcus, Tristan Harris, Nobel laureates Oliver Hart and Sheldon Glashow, and Fields medalist Timothy Gowers. The list mixes AI safety researchers, open-source founders, accelerationists, scientists and legal scholars.
Who criticized the MisAnthropic Statement?
Communication researcher Nirit Weiss-Blatt wrote that the signers “don't see the irony” of condemning Anthropic’s power while their own groups have sought sweeping restrictions. Mechanize co-founder Matthew Barnett wrote “I stand with Anthropic,” and co-founder Ege Erdil endorsed his reply.
Has Anthropic responded to the MisAnthropic Statement?
SnowRock had found no public response from Anthropic as of 04:11 UTC on November 19, 2026. Its IPO filing separately says it does not expect to be able to safely use Claude to fully autonomously build successive frontier models without meaningful human supervision.
What does “closing the loop” mean in Anthropic’s IPO filing?
Anthropic’s prospectus defines “closing the loop” as a cycle in which successive versions of its AI models help build the next. The filing calls it “one of the most significant structural advantages in frontier intelligence.”
What is recursive self-improvement, or RSI?
Recursive self-improvement is the idea of AI systems improving their own successors, speeding up AI progress with less human work. Anthropic’s filing says using AI to help build more capable AI “could also pose a considerable safety risk, particularly if it sharply accelerates the pace of progress,” while critics including Max Tegmark say Anthropic is racing toward it.
Who controls Anthropic after the IPO?
Anthropic’s seven co-founders hold 50.1% of the total vote through a single Class F share owned by their LLC, which carries no economic rights. A Long-Term Benefit Trust elects the Class T directors, who make up a majority of the board.
Is Anthropic profitable?
No. In the first nine months of 2026, Anthropic had a loss from operations of about $1.3 billion on $33.2 billion of revenue, and a net loss of about $104.0 billion, almost all of it from items outside its operations, according to its IPO filing.
Are Anthropic’s founders selling shares in the IPO?
Yes. Each of the seven co-founders, including Dario and Daniela Amodei, is selling 3,378,378 shares, about $280 million apiece at $83 and about $1.96 billion in all.
How much revenue does Anthropic make?
Anthropic reported $33.2 billion of revenue for the first nine months of 2026, up from $2.5 billion a year earlier, and a loss from operations of about $1.3 billion. It put its annualized run-rate revenue at about $81 billion as of October 31, 2026.



