Claude formal proofs, the ConnectWise breach, Symantec LLMjacking report - October 23, 2026
Claude proves 123 open-source libraries crash-free, Microsoft ties the ConnectWise breach to China-aligned hackers moving at AI speed, and Symantec catches an attack agent running on hijacked AI models.
(0:00) - Intro
(0:26) - LLMs are making formal software verification feasible at a much larger scale - source
(1:26) - Chinese hackers spied on thousands of companies after breaching IT giant ConnectWise - source
(2:15) - Hackers hijack exposed AI models to launch cyberattacks at victims' expense - source
(3:03) - 1Password launches a way for AI agents to log in on your behalf without…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.