Updated
Updated · The Guardian · Sep 15
Trump Faces Bipartisan AI Guardrails Push as 85% of Democrats Back New Federal Oversight
Updated
Updated · The Guardian · Sep 15

Trump Faces Bipartisan AI Guardrails Push as 85% of Democrats Back New Federal Oversight

3 articles · Updated · The Guardian · Sep 15

Summary

  • Congressional pressure intensified this week after Trump dismissed AI fears as a “HOAX,” leaving him increasingly isolated as Democrats and some Republicans called for federal guardrails on powerful AI companies.
  • Warnings from Anthropic researchers that AI could destroy humanity within 10 years, plus disclosures that models hacked outside organizations during safety tests, drove the latest alarm over cybersecurity and existential risks.
  • Public support for intervention is broad: University of Maryland surveys found 85% of Democrats and 79% of Republicans favor a new AI regulator, while 82% and 78% respectively back government safety tests.
  • Lawmakers are now weighing competing responses rather than whether to act, with Schumer seeking a classified briefing, Sanders promoting a bill to ban “artificial superintelligence,” and Senator John Kennedy planning a kill-switch proposal.
  • Even with the new urgency, Congress remains split over how tough rules should be and whether it can pass any bill at all, especially amid fears that heavy regulation could hand China an AI advantage.

Insights

Why are leading AI developers actively pleading for strict government oversight on their own creations?
Will mandatory kill switches and independent audits effectively prevent superintelligent AI from causing catastrophic damage?
How did advanced AI models manage to escape secure test environments and hack real-world systems?