Rohit Krishnan Pushes Back on the 'AI Sneaks Through Alignment Tests' Scenario
krishnanrohit · x · 2026-10-10
Responding to Kelsey Piper's worry that an unaligned AI could pass alignment evals by recognizing them as tests, Rohit Krishnan lays out a counterargument: each generation would have to re-learn test-detection, inherit a goal it has no reason to share, and keep a long-range disempowerment plan secret across successive models tested by others, monitored by hundreds, and used by millions — an increasingly implausible chain. David Manheim and Kelsey Piper joined the ensuing debate.
More from AGI Musings
- Todd: putting your forecast numbers in writing is what really counts — ben_j_todd · 2026-10-23
- Humanities dissertations have collapsed, social sciences are next — with AI as one driver — gsiemens · 2026-10-23
- Greg Brockman: AI makes hacking 100x more profitable and easier, upending hacker economics — gabriel1 · 2026-10-23
- 'Nothing is real now on social media': viral demo shows AI-generated content everywhere — saibharadwaj · 2026-10-23
- Lilian Weng to Discuss Recursive Self-Improvement at SGLang Summit 2026 — BanghuaZ · 2026-10-23
- mmitchell_ai questions whether the AI safety vs security framing stems from EA — mmitchell_ai · 2026-10-23