The ChatGPT maker says its upcoming Astra related model term may have reached “critical” cyber capabilities, prompting it to halt a significant number of training alternative runs while it tightens internal safeguards.
The ChatGPT maker says its upcoming Astra related model term may have reached “critical” cyber capabilities, prompting it to halt a significant number of training alternative runs while it tightens internal safeguards.
Source: https://www.wired.com/story/openai-overhauls-safety-protocols-after-its-ai-agents-went-rogue/
