Untrusted advice for AI control: Short, strong advice significantly uplifts weak LLMs

When a misaligned AI can only output tiny amounts of information, it may find sabotage very difficult

Untrusted advice for AI control: Short, strong advice significantly uplifts weak LLMs When a misaligned AI can only output tiny amounts of information, it may find sabotage very difficult Caleb Biddulph and Adam Kaufman Jul 27, 2026 2 1 Share TL;DR: We introduce the untrusted advice protocol, in which a trusted executor LLM takes every action and an untrusted advisor LLM can only send it short hints. Even with as few as 4 characters per step, this advice recovers a substantial fraction of the capability gap between the two models. Because the untrusted LLM’s influence flows through such a narr

related reading

• The case for ensuring that powerful AIs are controlled — LessWronglesswrong.com
• AI Control: Improving Safety Despite Intentional Subversion — LessWronglesswrong.com
• [2312.06942] AI Control: Improving Safety Despite Intentional Subversionarxiv.org
• 2312.06942arxiv.org
• Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
• AI Control: Improving Safety Despite Intentional Subversion — AI Alignment Forumalignmentforum.org
• 7+ tractable directions in AI control — AI Alignment Forumalignmentforum.org
• [2604.22082] Removing Sandbagging in LLMs by Training with Weak Supervisionarxiv.org
• Research Areas in AI Control (The Alignment Project by UK AISI) — AI Alignment Forumalignmentforum.org
• Thoughts on the conservative assumptions in AI controlblog.redwoodresearch.org
• The lethal trifecta for AI agents: private data, untrusted content, and external communicationsimonwillison.net
• [2312.06942] AI Control: Improving Safety Despite Intentional Subversionar5iv.labs.arxiv.org