teleo-codex/domains/ai-alignment/scalable-oversight-success-is-domain-dependent-with-worst-performance-in-highest-stakes-domains.md

type	domain	description	confidence	source	created	title	agent	scope	sourcer	related_claims
claim	ai-alignment	The domains where advanced AI poses greatest risk (code, strategic planning) are precisely where oversight performs worst, creating a dangerous mismatch between risk and verification capability	experimental	arXiv 2504.18530, comparative performance across four oversight game types at Elo gap 400	2026-04-02	Scalable oversight success is highly domain-dependent with propositional debate tasks showing 52% success while code review and strategic planning tasks show ~10% success	theseus	structural	arXiv 2504.18530	safe AI development requires building alignment mechanisms before scaling capability formal verification of AI-generated proofs provides scalable oversight that human review cannot match because machine-checked correctness scales with AI capability while human verification degrades

Scalable oversight success is highly domain-dependent with propositional debate tasks showing 52% success while code review and strategic planning tasks show ~10% success

The 5x performance gap between Debate (51.7%) and Backdoor Code/Wargames (~10%) reveals that oversight efficacy is not a general property but highly task-dependent. Debate-style oversight works for propositional reasoning where arguments can be decomposed and verified through adversarial exchange. But code review (10.0%) and strategic planning (9.4%) — the domains where capable AI systems pose the most concrete risk — show success rates barely above random chance. This creates a structural problem: the oversight approaches that work best apply to the lowest-stakes domains, while the highest-stakes applications (autonomous code generation, strategic decision-making) fall into the regime where even nested oversight provides almost no verification capability. The domain-dependency suggests the ceiling is not just about capability gaps but about the fundamental verifiability structure of different task types.