Every AI control attempt is a tighter cage, argues Nina Miolane

ninamiolane · x · 2026-09-21

Nina Miolane argues that every attempt to control AI since the sandbox-escape incident has been a tighter cage: stricter isolation, a kill switch, or a second AI watching the first. These measures are worth taking, but they expose that we cannot yet predict what AI will do — we're only preparing to react. It's part of her Science of Intelligence discussion series.

Original post →

More from AGI Musings

AGI Musings channel →