Auditing language models for hidden objectives \ Anthropic

A collaboration between Anthropic's Alignment Science and Interpretability teams

Alignment Interpretability Auditing language models for hidden objectives Mar 13, 2025 Read the paper A new paper from the Anthropic Alignment Science and Interpretability teams studies alignment audits —systematic investigations into whether models are pursuing hidden objectives. We practice alignment audits by deliberately training a language model with a hidden misaligned objective and asking teams of blinded researchers to investigate it. This exercise built practical experience conducting alignment audits and served as a testbed for developing auditing techniques for future study. In King

related reading

• Auditing language models for hidden objectives — LessWronglesswrong.com
• [2503.10965] Auditing language models for hidden objectivesarxiv.org
• [2602.22755] AuditBench: Evaluating Alignment Auditing Techniques on Models with Hidden Behaviorsarxiv.org
• Building and evaluating alignment auditing agents — AI Alignment Forumalignmentforum.org
• Natural Language Autoencoders \ Anthropicanthropic.com
• AuditBenchalignment.anthropic.com
• [2602.22755] AuditBench: Evaluating Alignment Auditing Techniques on Models with Hidden Behaviorsarxiv.org
• Building and evaluating alignment auditing agentsalignment.anthropic.com
• [2604.16812] Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
• A Pragmatic Vision for Interpretability — AI Alignment Forumalignmentforum.org
• How confessions can keep language models honest | OpenAIopenai.com
• Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org