Was I wrong about open models & cyber risks?
As AI models actively begin discovering critical zero-day vulnerabilities in the wild, this piece examines the real-world offense-defense balance and tests long-standing predictions about the cybersecurity risks of open-weight models.
What happened
The author re-evaluates his earlier claim that open-weight AI models like GLM-5.3 and a hypothetical open-weight Claude Mythos would not cause catastrophic cybersecurity harm, following a wave of AI-driven vulnerability disclosures dubbed "Black October." While acknowledging a sharp rise in critical exploits and strained cyber defense teams, he argues the situation represents an expensive industry transition and acceleration of existing trends rather than societal collapse, with closed models currently playing a larger role than open ones.
Who it’s about
- Claude Mythos
- GLM-5.3
- Anthropic
- Epoch AI
- Interconnects
Source
Interconnects
Was I wrong about open models & cyber risks?
More on security
Get a brief picked for you
Sign in to follow topics and companies, save stories and search the archive.
Sign in