Bilal Chughtai left DeepMind’s AGI safety team in July and posted his warning this week

Amid escalating debates over artificial intelligence safety, Washington lawmakers and key defense officials remain divided on how aggressively to govern tech platforms. In the Senate, lawmakers Thune, Cruz, and Klobuchar are currently negotiating legislation designed to mandate a duty of care for software creators to prevent catastrophic risk. Conversely, Donald Trump explicitly rejected the alarming claims last week and reiterated his skepticism during a Monday phone conversation with Jensen Huang. Meanwhile, Emil Michael, the Pentagon’s chief technology officer, categorized the anxiety as a doom loop, arguing that market dynamics and government involvement can successfully address dangers without halting progress.

These political maneuvers follow a surge of warnings from technical insiders, including Anthropic’s Evan Hubinger, who cited internal estimates placing the existential threat this decade above 1

Leave a Reply

Your email address will not be published. Required fields are marked *