Person
Eliezer Yudkowsky
Cited in eleven posts, among them AI Alignment with Changing and Influenceable… and Existential Risk from Power-Seeking AI, since September 2026
Where they come up
Said by someone else — these sentences name them.
“However, the robustness of this hope is challenged by the nearest unblocked strategy problem (Yudkowsky, 2015) : the problem that an AI which strongly optimizes for a (misaligned) goal will exploit even small loopholes in (aligned) constraints, which may lead to arbitrarily bad outcomes (Zhuang and Hadfield-Menell, 2020) .”
· The Alignment Problem from a Deep Learning Perspective · machine-resolved
“Thanks to Eliezer Yudkowsky, Rohin Shah, and Evan Hubinger for comments on the relevant scope here (which isn’t to say they endorse my choice of definition).”
· Existential Risk from Power-Seeking AI · machine-resolved
In the claim ledger
3 promoted claims about them. Assessments are the model’s knowledge, not verification.
Learned aligned constraints cannot be assumed loophole-free against a strongly optimizing misaligned goal.
plausible · The Alignment Problem from a Deep Learning Perspective
Placing a human in an idealized state to elicit unbiased feedback is judged even less feasible than debiasing their feedback after the fact.
plausible · AI Alignment with Changing and Influenceable Reward Functions
The paper claims ideal observer theory and coherent extrapolated volition specify reward functions that cannot be reached within its framework, requiring an extension to unreachable rewards.
plausible · AI Alignment with Changing and Influenceable Reward Functions