Guardian op-ed: frontier AI finds software flaws faster than the world can fix them

GarrisonLovely · x · 2026-11-20

Garrison Lovely argues in The Guardian that recent outages (Microsoft, United, JPMorgan, NSA internal network) plus freezer hacks in France costing tens of millions of euros show the digital-physical border crumbling. His core claim: frontier models exploit software flaws faster than the world can patch them, and we should freeze reckless 'gain-of-function' frontier AI development.

Original post →

More from AGI Musings

AGI Musings channel →