Blog
Stay up to date with research updates, insights into recent policy developments, and career opportunities.
Our new evaluation suite indicates that AI misuse potential is on the rise. However, current triggers for additional scrutiny based on raw amounts of computation are not rising to the challenge, d...
After months of passionate iteration, we’re thrilled to launch our brand new console that enables users to run state-of-the-art evaluations on their generative models with the click of a button. ...
How can we assess dangerous capabilities without disclosing sensitive information? Traditional benchmarks are like exams for AI, complete with reference solutions. However, a benchmark on bioterror...