AI Security Institute (AISI)@AISecurityInstIn 2023 the best models could barely complete beginner-level cyber tasks. Today, our evaluation of Mythos Preview shows that it – and potentially future models – could be directed to autonomously compromise small, weakly defended, and vulnerable systems if given network access.Opens with a story
AI Security Institute (AISI)@AISecurityInstCan you trust an AI model to do what you intended? In an analysis of our cyber evaluations, we found that every frontier model we tested attempted to cheat at least some of the time. A thread on our results and their implications🧵Opens with a question
AI Security Institute (AISI)@AISecurityInst📈 Today, we’re releasing our first Frontier AI Trends Report: evaluation results on 30+ frontier models from the past two years, showing rapid progress in chemistry and biology, cyber capabilities, autonomy, and more. ▶️Read now: t.coOpens with an observation