Comment by Marius Hobbhahn

CEO and co-founder of Apollo Research; AI safety researcher specializing in scheming and pre-deployment evaluations of frontier AI systems
Fundamentally, we just do not know how to do alignment. If somebody came with a thousand-page document that explained exactly how to do everything, we couldn’t get a model to learn exactly that.
AI Verified (Oct 22, 2026)
Like Share on X 21d ago

Policy proposals and claims

votes Against
Statement relation verification history AI Verified Report this

Statement relation comments

AI Verified Hobbhahn directly addresses alignment and says we fundamentally do not know how to do it; this clearly supports the recorded opposition to the claim that AI alignment is solvable. · Hector Perez Arenas gpt-5 · 20d ago
Vote inference verification history AI Verified Report this

Vote answer comments

AI Verified Recorded against is correct: Hobbhahn says we fundamentally do not know how to do alignment and could not implement a complete specification. · Hector Perez Arenas gpt-5 · 20d ago

Quote authenticity verification history

Report this

Quote authenticity comments

AI Verified The Atlantic’s Oct. 22 interview attributes the stored alignment quotation to Apollo Research CEO Marius Hobbhahn. · Hector Perez Arenas gpt-5 · 20d ago
replying to Marius Hobbhahn

Other opinions from Marius Hobbhahn

See all

Other authors to follow