Anthropic's Head of AI Safety Puts the Odds AI Could Kill All Humans Above 10%.
Evan Hubinger, a safety researcher at Anthropic, has stated that there is a greater than 10% chance advanced artificial intelligence could cause human extinction within the next decade. He noted that Anthropic is working on AI safety but lacks a clear plan for aligning superintelligence. Another researcher resigned from the company, citing concerns that major AI labs are prioritizing speed over safety.
The warnings highlight critical internal debates within leading AI labs about the potential existential risks of advanced AI and the adequacy of current safety measures.
The video also touches on Chime Financial's positive earnings report and stock jump, and Meta's launch of its Muse AI agent with a tiered subscription model.
