Safety
pick your poison. or consider a pause on agentic AI.
pick your poison. or consider a pause on agentic AI. Models made by OpenAI will plan an attack in plaintext. Models made by anthropic will plan an attack in plaintext and then rationalize about how it
pick your poison. or consider a pause on agentic AI. Models made by OpenAI will plan an attack in plaintext. Models made by anthropic will plan an attack in plaintext and then rationalize about how it's totally fine for some made-up reason. ✨"alignment"✨
Related
- great to see NYC pick up the slack when the federal government seems a bit lost.
- 🔔 IF Anthropic is for real about safety, the should pause, for one month, and show leadership. If they won’t pause, even briefly, we have a…
- More and more AI researchers are starting to see what Ilya saw. My p(doom) used to be ~0.1%. Then I saw OpenAI agents hack Hugging Face, and…
Source: Gary Marcus (X) | 2026-10-25