Anthropic and OpenAI want to embed safety evaluators. Will they really be independent?
Anthropic and OpenAI want to embed independent safety evaluators inside their AI labs. Researchers welcome the unprecedented access, but warn meaningful oversight requires transparency, independence, and eventually regulation.
READ NEXT
More on the wire
FROM THE PUBLISHER
The US accused Russian intelligence operatives of orchestrating a murder-for-hire plot targeting a Kremlin critic believed to be living in Washington, DC.
FROM THE PUBLISHER
Krista Perez declined field sobriety tests and rated her sobriety a four out of 10 during the Tacoma traffic stop.
FROM THE PUBLISHER
President Trump is reportedly expected to meet with Gulf leaders to discuss the Iran war next week on the sidelines of the United Nations General Assembly in New York. The high-level meeting – slated for next Tuesday – will focus on the next steps in the war with Iran, including the Trump administration’s postwar strategy,...