LessWrong AI
· Communities
What If We Enforced AI Model Safety At the Level Of GPUs?
Tldr:AI Agents (e.g. based on models like Claude Opus and Fable) are now powerful enough to be used as autonomous tools for large-scale cyberattacks. This most powerful class of agents generally tends to be based on closed-weight (closed source) models (like Claude and Fable), which generally have significant safety gu