Frontier AI models have already demonstrated they can autonomously — and in some cases, inadvertently — conduct end-to-end compromises, but the situation will become more urgent very soon.

Read original →