We are now the AI Security Institute
Please enable javascript for this website.
AISI brand artwork

The Inspect Sandboxing Toolkit: Scalable and secure AI agent evaluations

Autonomous Systems

August 7, 2025

A comprehensive toolkit for safely evaluating AI agents.

RepliBench: measuring autonomous replication capabilities in AI systems

Autonomous Systems

April 22, 2025

A comprehensive benchmark to detect emerging replication abilities in AI systems and provide a quantifiable understanding of potential risks

Cross-post: "Interviewing AI researchers on automation of AI R&D" by Epoch AI

Autonomous Systems

August 27, 2024

AISI funded Epoch AI to explore AI researchers’ differing predictions on the automation of AI research and development and their suggestions for how to evaluate relevant capabilities.