Principles for Meta-Science and AI Safety Replications — LessWrong
lesswrong.com · 2,368 words · saved by 1 readers
If we get AI safety research wrong, we may not get a second chance. But despite the stakes being so high, there has been no effort to systematically…
x Principles for Meta-Science and AI Safety Replications — LessWrong Replication Crisis AI Frontpage 48 Principles for Meta-Science and AI Safety Replications by Zephaniah Roe 23rd Jan 2026 4 min read 7 48 If we get AI safety research wrong, we may not get a second chance. But despite the stakes being so high, there has been no effort to systematically review and verify empirical AI safety papers. I would like to change that. Today I sent in funding applications to found a team of researchers dedicated to replicating AI safety work. But what exactly should we aim to accomplish? What should AI
related reading
- Tips for Empirical Alignment Research — AI Alignment Forumalignmentforum.org
- The case for ensuring that powerful AIs are controlled — LessWronglesswrong.com
- The Case Against AI Control Research — LessWronglesswrong.com
- Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
- AI Control: Improving Safety Despite Intentional Subversion — LessWronglesswrong.com
- AI safety undervalues founders — LessWronglesswrong.com
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- Core views on AI safety: When, why, what, and how \ Anthropicanthropic.com
- How do we (more) safely defer to AIs? — LessWronglesswrong.com
- AI Safety Seems Hard to Measurecold-takes.com
- How might we safely pass the buck to AI? — LessWronglesswrong.com
- Ten AI safety projects I'd like people to work onthirdthing.ai