Skip to main content

CAIS joins statement challenging Anthropic’s safety branding ahead of IPO

20 NOVEMBER 2026·2 MIN READ·5 SOURCES

The Center for AI Safety (CAIS) said it joined a coalition signing the MisAnthropic Statement ahead of Anthropic’s upcoming IPO. The statement challenges Anthropic’s safety branding and accuses the company of pursuing recursively self-improving AI while seeking to consolidate power.

Key takeaways · 4

  • 01

    CAIS said it joined the coalition behind the MisAnthropic Statement ahead of Anthropic’s IPO.

  • 02

    The statement accuses Anthropic of using safety branding to court investors, employees and the public.

  • 03

    CAIS says Anthropic weakened its Responsible Scaling Policy in February, including by abandoning unilateral commitments to pause.

  • 04

    Anthropic priced its IPO at $83 per share, above the previously indicated $70–$78 range, according to The Hindu.

What the statement argues

CAIS said the statement was published on misanthropicstatement.com and describes a disconnect between Anthropic’s safety branding and its effort to build recursively self-improving AI and consolidate power.[1] The signatories accuse Anthropic of racing to build such systems despite warning that they could concentrate power and escape human control.[2] They also say the company uses safety branding to court investors, employees and the public.[2] The statement calls Anthropic “not a safe AI company” and urges rejection of what it characterizes as a reckless attempt to amass power.[2]

A coalition with varied views

CAIS described the signatories as including AI safety researchers, open-source advocates, AI ethicists and accelerationists.[1] The Hindu reported that technology signatories include Hugging Face co-founder Thomas Wolf and xAI co-founder Igor Babuschkin; it also named physicist Max Tegmark, CAIS executive director Dan Hendrycks and accelerationist Guillaume Verdon.[3] CAIS said coalition members disagree on almost everything but share concern about a company treating its own dominance as the greater good.[1] The coalition’s shared criticism, according to CAIS, is Anthropic’s effort to cast itself as the industry’s “good guys.”[1]

Specific criticisms from CAIS

CAIS said Anthropic weakened its Responsible Scaling Policy in February, including by abandoning unilateral commitments to pause.[1] The statement also alleges that Anthropic quietly weakened its safety pledges and lobbied for regulation it considers self-serving.[2] In a quoted passage, CAIS says the company lobbied to weaken SB 1047, the RAISE Act and the AEGIS Act, and privately advocated for federal pre-emption of state AI laws.[1] The statement further argues that recursive self-improvement was presented to public-market investors as a core competitive advantage.[1] These are claims made by CAIS and the signatories, not independently established findings in the cited material.

IPO details and response

Anthropic priced its IPO at $83 per share, above its previously indicated $70–$78 range, The Hindu reported.[4][3] Anthropic said the offering comprises 980 million Class A shares and that trading was expected to begin on Nasdaq’s Global Select Market on November 19, 2026, under the symbol ANT.[4] The company said it is selling 956,351,354 shares, while selling stockholders are offering 23,648,646 shares; Anthropic said it would receive no proceeds from the stockholders’ sales.[4] The Next Web reported that Anthropic had not publicly responded to the statement by Thursday morning.[5] Its amended IPO filing said Claude’s successors could not safely be built without meaningful human supervision.[5]

The statement puts Anthropic’s safety branding and pursuit of self-improving systems into public debate. Treat the coalition’s assertions as allegations, and distinguish them from Anthropic’s amended filing statement that Claude’s successors could not safely be built without meaningful human supervision.

Why it matters
For teams

AI training for your team

kju.ai is built for teams: daily sessions, team analytics, and rollout support for your organization.

How this developed

  1. 20 November 2026

    CAIS joins statement challenging Anthropic’s safety branding ahead of IPO

Sources

AI fluency, one session a day, built for your work.