Why we are excited about confession! — LessWrong

Boaz Barak, Gabriel Wu, Jeremy Chen, Manas Joglekar …

x Why we are excited about confession! — LessWrong AI Curated 2026 Top Fifty: 11 % 140 Why we are excited about confession! by Boaz Barak , Gabriel Wu , Manas Joglekar 14th Jan 2026 AI Alignment Forum Linkpost for alignment.openai.com 10 min read 32 140 Ω 54 Boaz Barak, Gabriel Wu, Jeremy Chen, Manas Joglekar [Linkposting from the OpenAI alignment blog , where we post more speculative/technical/informal results and thoughts on safety and alignment.] TL;DR We go into more details and some follow up results from our paper on confessions (see the original blog post ). We give deeper analysis of t

related reading

• Why We Are Excited About Confessionsalignment.openai.com
• How confessions can keep language models honest | OpenAIopenai.com
• confessions_paper.pdfcdn.openai.com
• Alignment remains a hard, unsolved problem — LessWronglesswrong.com
• Deep Deceptiveness — LessWronglesswrong.com
• AI Control: Improving Safety Despite Intentional Subversion — LessWronglesswrong.com
• Should We Train Against (CoT) Monitors? — LessWronglesswrong.com
• Simulated Users & Sad LLMs1a3orn.com
• Self-Fulfilling Misalignment Data Might Be Poisoning Our AI Modelsturntrout.com
• Training on Documents about Reward Hacking Induces Reward Hackingalignment.anthropic.com
• The Most Forbidden Technique — LessWronglesswrong.com
• Natural-emergent-misalignment-from-reward-hacking-paper.pdfassets.anthropic.com