Europe’s new AI rules elicit techie pledges as the US leaves oversight to the states

Regulators, mount up… there’s AI to wrangle. The EU’s AI Act, which went into effect in August, applies new rules to so-called high-risk artificial intelligence (like: education, law enforcement) and bans AI for social-scoring and predictive policing. Still, most of its provisions won’t hit until 2026 — leaving room for tech titans to feel things out. Microsoft and Google signed a pledge saying they’d get a head start on following the regs, but Meta — whose open-source Llama model is intended to be used freely by developers — kicked the can.

Instructions unclear… The US has made its own efforts to rein in AI. Last year President Biden issued an executive order that sought to study AI’s effect on the labor market and require AI safety assessments. Still, there’s been no federal AI legislation, leaving it up to the states. That’s had mixed results: California’s Gov. Gavin Newsom vetoed an AI-regulation bill last month, while states like Colorado and Illinois have passed limited regs. At least 26 states have passed rules around genAI in election communications.

AI on the polls… As federal legislators work to pass AI laws — Senate Majority Leader Chuck Schumer said he hoped it could happen by the end of the year — both former President Trump and VP Harris have promised to promote some form of AI policy if elected president.