5 citations · 6 across the 5 of their papers we have counts for
6 papers
Red-Teaming Multi-Agent Attacks for Loss-of-Control Risks
John Kitaoka, Thao Amelia Pham, Carlos Rafael Giudice +1
AI agents are growing more capable and persistent, and are being deployed in ever larger numbers. As more start collaborating, AI control measures must remain effective when they c…
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…
UK AISI Alignment Evaluation Case-Study
Alexandra Souly, Robert Kirk, Jacob Merizian +2
This technical report presents methods developed by the UK AI Security Institute for assessing whether advanced AI systems reliably follow intended goals. Specifically, we evaluate…
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…
Establishing Best Practices for Building Rigorous Agentic Benchmarks
Yuxuan Zhu, Tengjun Jin, Yada Pruksachatkun +22
Benchmarks are essential for quantitatively tracking progress in AI. As AI agents become increasingly capable, researchers and practitioners have introduced agentic benchmarks to e…
Interpreting Neural Networks through the Polytope Lens
Sid Black, Lee Sharkey, Leo Grinsztajn +8
Mechanistic interpretability aims to explain what a neural network has learned at a nuts-and-bolts level. What are the fundamental primitives of neural network representations? Pre…