November 4, 2024
October 21, 2024
Some rough ideas
October 15, 2024
Some reactions
September 23, 2024
Testing an LM system for dangerous capabilities is crucial for assessing its risks
July 27, 2024
July 10, 2024
And new-ish page: Policy advocacy
June 12, 2024
New details on the Long-Term Benefit Trust, but most questions remain
May 29, 2024
May 27, 2024
Anthropic should share the details
May 24, 2024
But they should
May 21, 2024
16 companies commit to make RSPs
May 17, 2024
But they are doing model evals for dangerous capabilities