AI Leaders Clash Over Safety and Regulation

A growing dispute over AI safety pits warnings from executives and researchers against technology leaders who oppose broad development slowdowns or new regulation. Anthropic CEO Dario Amodei, OpenAI leaders and others have called for pauses or stronger safeguards, including independent evaluators and shared standards, while Geoffrey Hinton has warned that AI could pose a serious threat to humanity; Andrew Ng and Marcel Salathé have dismissed extinction-focused predictions as speculative. Elon Musk has proposed that leading AI companies, including OpenAI, Anthropic, Google, Meta and major Chinese firms, let rivals test their models through a shared test harness before public release, but no laboratory has committed to the plan. Musk said peer evaluations could identify safety problems missed by internal assessments, while Anthropic researcher Evan Hubinger estimated a more than 10% chance that AI could kill all humans within the next decade and former Anthropic researcher Jacob Coxon accused leading labs of gambling with lives. Mark Zuckerberg and Nvidia CEO Jensen Huang say market and legal incentives encourage companies to build aligned systems safely, though Zuckerberg supports independent advisers and Meta delayed its Muse AI rollout for safety work; reports also say Musk, Zuckerberg and Huang helped persuade President Donald Trump to reject an industry-funded AI regulator. In Congress, Sen. Rand Paul blocked a proposed government kill-switch bill as lawmakers continued debating a broader duty-of-care framework for AI developers.