3 ms·
The core details on what they consider dangerous are here: https://www.anthropic.com/news/core-views-on-ai-safety https://www.anthropic.com/news/core-views-on-a
by muzani 2y ago
The core details on what they consider dangerous are here: https://www.anthropic.com/news/core-views-on-ai-safety https://www.anthropic.com/news/core-views-on-ai-safety
The linked article seems to be a much lower level on the implementation details.