David Duvenaud
@DavidDuvenaud
This is imo the most sophisticated attempt so far to build a robust safety case regarding novel dangers on a frontier model. It's valuable for public discourse that Anthropic is publicizing the current arguments, sources of evidence, and third-party criticisms.
Sam Bowman@sleepinyourhat · Oct 28, 2025🌱⚠️ weeds-ey but important milestone ⚠️🌱
This is a first concrete example of the kind of analysis, reporting, and accountability that we’re aiming for as part of our Responsible Scaling Policy commitments on misalignment.
This is a first concrete example of the kind of analysis, reporting, and accountability that we’re aiming for as part of our Responsible Scaling Policy commitments on misalignment.
Open quoted post →
1 45