New
InnovationMore in Innovation→
Anthropic’s Amanda Askell on the Ethics of Agentic AI
Key Takeaways
- Anthropic is reworking AI safety around systems that act autonomously over longer time horizons.
- Amanda Askell distinguishes between asking an AI for ethical advice and letting it make decisions for a user.
- Anthropic’s evolving constitution is central to how it guides Claude’s behavior as capabilities expand.
DE
DT Editorial Team··via fastcompany.com