Philosopher Seth Lazar rebuts 'alignment is impossible' essay as facile cliché
sethlazar · x · 2026-09-13
Responding to the essay 'AI Alignment Is Impossible', philosopher Seth Lazar calls it a facile and hackneyed observation.
- Lazar argues that 'solving alignment' is a precise technical goal for computer scientists: designing systems that robustly adhere to the reasons that apply to them—something that can clearly be done better or worse.
- He distinguishes the specification of values from their technical implementation, a distinction standard since at least 2020; the value-content question was never going to be settled definitively, which doesn't make alignment impossible.
- He adds that catchphrases like 'aligned to whom' suggest the author hasn't engaged with the field.
Related event: Philosophers Clash Over Whether AI Alignment Is Impossible(2 posts)→
More from AGI Musings
- Ex-DeepMind synthetic biologist: AI biorisk pandemic claims are "straight out of the crackpipe" — jeremyphoward · 2026-09-14
- Deepak Nathan: Today's Models Are Too Dumb at Ambiguity, Not Too Smart — deepakns · 2026-09-14
- Alex Irpan revisits Gwern's classic essay on why Tool AIs lose to Agent AIs — AlexIrpan · 2026-09-14
- Melanie Mitchell pushes back on 'rogue AI swarm' narrative as misleading metaphor — anilkseth · 2026-09-14
- Models aren't too smart — they're too dumb in the face of ambiguity; guardrails matter more — fooobar · 2026-09-14
- Two-year-old AI podcast predictions largely played out as expected — misovalko · 2026-09-14