Analysis: recent open weight models lag frontier closed models' cyber capabilities by 4 to 7 months, a narrower gap than the 6 to 10 months through most of 2025 (AI Security Institute)

AI Security Institute:
Analysis: recent open weight models lag frontier closed models' cyber capabilities by 4 to 7 months, a narrower gap than the 6 to 10 months through most of 2025  —  AISI has tracked the cyber capabilities of frontier AI models since 2023.  On our evaluations, the most capable models …

Posted from: this blog via Microsoft Power Automate.

Daily Deals