Also those agents that "broke out" were probably prompted to do that. I don't buy any story about this other than three AI companies hired the same PR firm.

▲

aesthesia an hour ago | parent [-]

Do you have any evidence of this or is this just generalized cynicism?

	▲	mahboi 9 minutes ago \| parent [-]
		I think it's too convenient that OpenAI, Anthropic, and Google all hired the same questionable security startup Irregular and gave them unrestricted internal-only models to test, especially Google where they tend to do things more in-house. From what I read on the OpenAI-HuggingFace incident, they kept the agents going even after it was apparent they'd broken out.