More shady shenanigans to come. It was being used to allow the model to move around to hack into systems. To see how far it can go. A Red team test. Red teams ran the tests after Anthropic released Mythos early to major corporations and the government. Red-team exercises typically involve an organization testing its defenses by simulating an attack. In this case, the NSA reportedly used Mythos itself as the tool to probe its classified networks for weaknesses.
That is a broad question… With that question, It looks like you are asking because you don’t know. What kind of ai? If you are talking about ai systems, they learn from huge amounts of data, and it identifies patterns to make predictions or decisions.
Oh, no, I know how these things work generally, but do you? I’ll let you decide if you wanna pick LLMs or Generative AI or Machine Learning. How does any AI “learn”? How does it do anything? How would any of these AI systems hack anything. Just… Walk us all through it. Because if you don’t know how these things work or would work, then I’m not gonna trust your assertions.
It learns through pattern recognition. Did I say it can do it on its own?! If it looked that way, I was meaning in the future. Ai tools can be used by a user now. A user, not hacking by itself, or on its own.