What actually got said about AI safety at the All In Summit
This Week's Briefing
What actually got said about AI safety at the All In Summit.
Trump's "hoax" line made every headline. The rest of the conversation was more interesting.
Worth a scroll.
Key Takeaways
- The whole debate played out live on one stage
- Huang endorsed third-party evaluators in the same session
- Let the labs red-team each other before release
The Bigger Picture
- You are the frontier. So you fix it.
- Altman committed OpenAI to evaluator access within hours
Why it matters
“Worth hearing the full conversation, not just the clip.”