4 papers
Multi-Agent AI Control: Distributed Attacks Hamper Per-Instance Monitors
Oliver Makins, Orazio Angelini, Zohreh Shams +1
AI control is a family of techniques to prevent an AI with malicious goals from subverting its operator's intent. AI Control usually studies a single agent in one trajectory, but r…
Propensity Inference: Environmental Contributors to LLM Behaviour
Olli Järviniemi, Oliver Makins, Jacob Merizian +2
Motivated by loss of control risks from misaligned AI systems, we develop and apply methods for measuring language models' propensity for unsanctioned behaviour. We contribute thre…
Async Control: Stress-testing Asynchronous Control Measures for LLM Agents
Asa Cooper Stickland, Jan Michelfeit, Arathi Mani +6
LLM-based software engineering agents are increasingly used in real-world development tasks, often with access to sensitive data or security-critical codebases. Such agents could i…
Auditing Games for Sandbagging
Jordan Taylor, Sid Black, Dillon Bowen +10
Future AI systems could conceal their capabilities ('sandbagging') during evaluations, potentially misleading developers and auditors. We stress-tested sandbagging detection techni…