Is AGI alignment even solvable in principle? Researcher questions the premise
danfaggella · x · 2026-10-19
Dan Faggella draws a provocative analogy: statements like "we should solve the evolution of all future species" or "solve civilization" are obviously silly because they aren't solvable problems. He then asks why we assume "we haven't yet solved AGI alignment" describes a problem that's solvable in principle at all — a fundamental challenge to the framing of alignment research.
More from AGI Musings
- willcb: AGI won't spike GDP — he sees just 3.5-5% growth by 2030 — willcb · 2026-10-19
- willcb debunks the 17% AGI growth forecast: robotics, UBI timelines don't hold up — willcb · 2026-10-19
- RL data startups get harder: insider says don't start one without lab-level alpha, as profit-sharing data shops skip VC — Scobleizer · 2026-10-19
- Matthew Green: six months into the "Mythos era", Teslas still get zero-click RCEs — the defender story isn't holding up — matthew_d_green · 2026-10-19
- 3Blue1Brown on the future of math: never a better time to be a mathematician — burny_tech · 2026-10-19
- Andrew Ng: optimism drives action, wishful thinking breeds dangerous complacency — andrewgwils · 2026-10-19