Orbit: A framework for multi-agent security evaluations

·LessWrong··

This post announces work completed as part of the MATS 9 program, supervised by Dr. Christian Schroeder de Witt. Moving forward, Orbit will be supported by the Cooperative AI Foundation. We're grateful to MATS and our Research Manager, Srija Chakraborty, for their support. Repo: https://github.com/wlanderson0/orbitWe’re excited to release v0 of Orbit, a framework for multi-agent safety and security evaluations built on Inspect. This is still a work in progress, and we expect to substantially upd...

Read full article →

Related Articles

LG TVs caught spying even when offline or on standby
sbulaev · Hacker News · 3h ago
Navier-Stokes – Tristan Buckmaster [pdf]
procedurecall · Hacker News · 14h ago
DHS 'Predictive Policing' Unit Is Analyzing Americans' Financial Habits
abraham · Hacker News · 5h ago
Google DeepMind Releases AlphaGenome Atlas
utiiiD · Hacker News · 4h ago
Benchmarking Qwen3.8 27B quantizations: 4-bit holds up, 1-bit collapses
stared · Hacker News · 4h ago