UN's First Independent AI Science Panel Warns the Window to Govern AI Is Closing

My take: This one had me thinking for a while. The UN released this week the preliminary report from its Independent International Scientific Panel on AI — 40 experts from around the world, co-chaired by Yoshua Bengio and Maria Ressa (Turing Award and Nobel Peace Prize, not small names). The headline that stopped me: they can't rule out catastrophic harm as AI grows more capable. But what really worries me isn't that — it's this: more than 40 AI governance frameworks already exist worldwide, and almost none are tested to see if they work, with many safety assessments done by the same companies building the AI. It's like having 40 different smoke detectors in 40 countries, with the manufacturer being the one who certifies they go off correctly 🔥. The panel feeds into the UN Global Dialogue on AI Governance opening July 6 in Geneva. The window to get this right is still open, they say. I hope so — because governments' 'we'll figure it out later' approach is going to get expensive.