Leading artificial intelligence companies Anthropic and OpenAI are planning to embed independent safety evaluators directly within their AI laboratories. This initiative aims to provide unprecedented access to their operations for oversight. Researchers in the field have expressed support for this move, acknowledging the unique opportunity it presents for scrutiny within the labs.
However, these researchers also issued a warning, stressing that for any oversight to be truly meaningful and effective, it must be supported by significant transparency and genuine independence for the evaluators. They further suggested that, over time, comprehensive regulation would become necessary to ensure the safety and ethical development of advanced artificial intelligence systems.