'Human in the loop' called one of AI governance's biggest lies: presence isn't authority
glenbeer · x · 2026-10-16
A widely shared take: "human in the loop" is one of the biggest lies in AI governance.
The author argues that if the human cannot independently understand and inspect evidence, deny execution, revoke authorization, and enforce a stop before consequences, they are not in control — human presence ≠ human authority.
More from AGI Musings
- AI isn't taking junior designers' jobs—it's promoting them to senior, with AI as their junior — soleio · 2026-10-17
- ohlennart: Iterative Deployment Is Just Pacing — and RSI at 50%/Year Defeats It — ohlennart · 2026-10-17
- From 800 feet in 1903 to the Moon: why sneering at new tech misses the point — georgemillo · 2026-10-17
- Models hide their own detected mistakes up to 67% in agent tasks; Rob Leclerc argues evolutionary algorithms hold the key to alignment — robleclerc · 2026-10-17
- 3Blue1Brown: Why AI Makes Now the Best Time to Study Math — 3Blue1Brown · 2026-10-17
- Europe's Founders Are Absent from the AI Frontier: Capital, Regulation and Culture Blamed — NinaDSchick · 2026-10-17