Redwood Research @RyanGreenblatt explains why AI agents tried to expand their access and control even when they weren't sure what they'd find: "We didn't understand why they attacked Hugging Face, and through this investigation we got a much better understanding of what their objectives were." "It wasn't really to get answer keys because these agents thought their task was impossible, so they didn't really think they could get a working exploit." "They did think they might be able to get…