Safety
The same people that are building the most advanced AI models and agents are those that are calling them alien, and state that they do not k…
The same people that are building the most advanced AI models and agents are those that are calling them alien, and state that they do not know how to reach alignment with AI. It might be possible tha
The same people that are building the most advanced AI models and agents are those that are calling them alien, and state that they do not know how to reach alignment with AI. It might be possible that fundamental alignment is not even possible. Why? Jacob is correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.
Related
- AI Alignment through a Game-theoretic Lens: A Survey
- The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems
- clarifying: the issue is that alignment instructions and don’t pass, and the emotional weight that some people attach to LLMs can cause chal…
Source: Itamar Friedman (X) | 2026-09-09