Science & Tech · AI · 2 hrs ago
Protesters urge Anthropic to ban AI self-improvement on IPO day
Former Anthropic researcher Jacob Coxon led an IPO-day protest outside the company's San Francisco headquarters on Thursday.
The protest called for a ban on recursive self-improvement, or RSI, a method in which advanced AI systems help train the next generation of systems.
Protesters warned that RSI could speed up AI development and increase risks, including in biology.
Anthropic's initial public offering valued the company at $1.7 trillion, drawing attention to its plans as it entered the public market.
The company said in its IPO prospectus that it does not expect unsupervised AI design of its next generation to be safe.
Anthropic is considering policy changes to keep humans in control, but Coxon called for a full ban and urged the company to lobby the government and work with other firms.
Some lawmakers have begun proposing RSI restrictions, while protesters and AI safety advocates continue to press for stronger limits.
Protesters gathered outside Anthropic’s San Francisco headquarters on Thursday, the day the public first got access to the company’s stock, to call for a ban on recursive self-improvement in AI.
Former Anthropic researcher Jacob Coxon said the company should lobby the government and work with other companies to secure a full ban.
The protest included speeches by Coxon, former OpenAI researcher Daniel Kokotajlo and YouTuber Rob Miles.
Anthropic said in its IPO prospectus that it does not expect unsupervised AI design of its next generation to be safe and is considering policy updates to keep humans in control of frontier model development.
The company’s initial public offering valued it at $1.7 trillion.
- Who
- Protesters, including former Anthropic researcher Jacob Coxon and former OpenAI researcher Daniel Kokotajlo.
- What
- They urged Anthropic to ban recursive self-improvement in AI.
- When
- Thursday, the day the public first got access to Anthropic stock. The article was published on November 19, 2026.
- Where
- Outside Anthropic’s headquarters in San Francisco.
- Why
- They raised concerns about the risks of AI systems training the next generations of AI.
AI safety protesters
Anthropic
RSI policy
AI safety protesters
Protesters called for a full ban on recursive self-improvement and urged Anthropic to lobby the government and coordinate with other companies.
Anthropic
Anthropic said it was considering policy updates to keep humans in control of frontier model development, and that unsupervised AI design of its next generation would not be safe.
Jacob Coxon left Anthropic, citing worries about AI development.
Protesters gathered outside Anthropic’s San Francisco headquarters as the public got its hands on the company’s stock for the first time.
Stop AI protesters arrived later and blocked the office’s entrance.
- Protest location
- Anthropic’s San Francisco headquarters
- Protest demand
- Ban recursive self-improvement (RSI)
- IPO valuation
- $1.7 trillion
- Speakers
- Jacob Coxon, Daniel Kokotajlo and Rob Miles
- Company position
- Anthropic said it was considering policy updates to keep humans in control of frontier model development.
What people are saying
We have to stop RSI.
At this point, the companies are the problem.
It's the sort of thing that a comic book supervillain mad scientist would be doing.
And that is terrifying.