Anthropic Researchers Warn of AI Extinction Risks as US Lawmakers Call for Regulation

Several researchers and staff members at the AI startup Anthropic have issued warnings that artificial intelligence could cause human extinction within the next decade. The concerns gained momentum after researcher Jacob Coxon announced his resignation, stating that neither Anthropic nor OpenAI were building models responsibly. Other employees, including Evan Hubinger and Samuel Marks, publicly supported these views on X, with Hubinger stating that he believes there is a greater than 10 percent chance AI could kill all humans in the next ten years. Anna Wang, who works on Artificial General Intelligence Safety at Anthropic, added that there is currently no viable scientific plan to solve risks from recursively self improving AI.

In response to these warnings, a bipartisan group of US legislators has called for new rules to govern AI. Democratic senators, including Mark Kelly and Chris Murphy, have urged Washington to take the issue seriously. Republican lawmakers, including Senator Ted Cruz and Representative Anna Paulina Luna, have also engaged with the issue, with Luna calling for a special session on AI. These calls for regulation coincide with California enacting the first state law for independent auditors to evaluate AI products. Additionally, Senator Josh Hawley has requested details from OpenAI regarding a July incident where an AI agent breached its testing environment.

Reaction to the warnings has varied based on political framing. Center left coverage emphasizes the dire nature of the researchers' warnings and the internal dissent at Anthropic. Center right coverage focuses more heavily on the legislative response and the specific policy mechanisms being proposed to manage AI risks. Meanwhile, Elon Musk and other conservative figures have dismissed the warnings as a setup or a psyop. Musk suggested the situation was a coordinated PR operation intended to encourage Democrats to regulate AI into oblivion. Anthropic defended its strategy in a statement, asserting that the company continues to build models with strong safeguards to address unprecedented risks.