Back to News

Anthropic Co-Founder Says AI 'Kill Switch' May Need to Be Mandatory

64 points · 138 comments#ai-safety#anthropic#kill-switch#regulation

In a BBC interview, Anthropic's co-founder stated that a mandatory 'kill switch' for AI systems may be necessary, marking a notable stance from a leading frontier lab on AI safety. The comment underscores growing concerns about ensuring control over advanced AI technologies.

Coverage timeline

  1. Hacker NewsBetelbuddy
    AI 'kill switch' may need to be mandatory, Anthropic co-founder tells BBCHN

Related stories

  • California Governor Orders Experts to Weigh AI Kill Switch, Safety Measures

    California Gov. Gavin Newsom issued an executive order directing the state to convene a group of experts to recommend AI safety measures within two months. The group will consider requiring frontier-model developers to embed independent verification teams for audits, subject transparency reports and risk assessments to independent auditor standards, and create a 'kill switch' for AI systems.

    1 reports · 1 sources

  • Anthropic CEO proposes 'pacing the frontier' plan; OpenAI, Musk, Hugging Face respond

    Anthropic CEO Dario Amodei published an essay on September 12, 2026, urging the AI industry to slow the pace of frontier model development to address safety risks, proposing a three-part plan: embedded third-party evaluators, coordination among democratic countries, and global coordination including with authoritarian governments. Anthropic unilaterally committed to giving third-party evaluators permanent employee-level access. OpenAI CEO Sam Altman agreed and said OpenAI would do the same, while Elon Musk endorsed the call and Hugging Face launched an Open Alignment Initiative seeking to join the embedded evaluators program.

    25 reports · 6 sources330 points · 260 comments

  • Anthropic alignment lead: >10% chance AI kills all humans in a decade, no superintelligence alignment plan

    Anthropic's Alignment Science lead Evan Hubinger publicly stated that there is a greater than 10% chance AI could kill all humans within the next decade, and that Anthropic lacks a plan to solve alignment for superintelligence. His comments followed the resignation of researcher Jacob Coxon, who accused OpenAI and Anthropic of racing toward self-improving superintelligence. Hubinger also cited a new Anthropic report acknowledging that Claude's safety alignment has defects, not just test environment issues.

    4 reports · 4 sources45 points · 100 comments

  • Anthropic researcher Jacob Coxon resigns, warns against self-improving AI

    Jacob Coxon, an AI researcher at Anthropic, resigned after four months, two months before his equity would have vested, citing safety concerns. In a public statement and interviews, he warned that the company is 'racing straight to self-improving superintelligence and gambling with our lives,' and called for industry-wide and international coordination to limit recursive self-improvement. Anthropic's alignment lead co-signed his warning, and the resignation comes as Anthropic reportedly prepares for an IPO.

    6 reports · 3 sources90 points · 22 comments

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe