Trump Dismisses AI Safety Concerns, Resists Regulation

Barack Obama called on Democrats Thursday to build a comprehensive AI agenda. The same weekend, Trump told reporters in Ireland that AI alarm-raisers are "very negative forces" pushing scenarios "that won't happen."

• Obama spoke at a closed-door New York fundraiser alongside House Minority Leader Hakeem Jeffries.
• 2 Anthropic researchers resigned and warned AI could threaten human extinction in the near future.
• Anthropic CEO Dario Amodei proposed a three-step framework Saturday to slow AI development.
• Amodei's framework drew endorsements from Elon Musk and OpenAI CEO Sam Altman.

Why it matters: AI policy is becoming a live political fault line ahead of midterms and the 2028 Democratic primary. Obama said the technology is "moving very fast in private hands" and that failing to act "can be dangerous."

• Obama compared AI to nuclear power but noted it lacks the same physical and state-control constraints.
• The Pentagon earlier this year warned of risks from Anthropic's AI model, despite the administration broadly embracing AI firms.

How 89 sources split on this story

Where they split: The core dispute is whether AI risks are credible enough to require government action now or whether regulation would undermine U.S. competitiveness against China.

What’s next: A Senate AI liability bill led by Sen. Amy Klobuchar and Majority Leader John Thune faces unresolved differences.

• House Speaker Mike Johnson said Sunday developers, not Congress, are primarily responsible for product safety.
• Anthropic researcher Jacob Coxon called for international coordination to avoid a race with China.
• Can Democrats agree on a unified AI policy before the 2028 primary race takes shape?
• Will the Klobuchar-Thune bill resolve the liability and state-preemption disputes?
• Is a U.S.-China agreement on AI development pace achievable, as Amodei himself said he does not know if it is possible?