Pinned
Everyone seems to have discovered recursive self-improvement this week.
Anthropic is warning about frontier AI systems improving themselves. OpenAI-adjacent investors are saying the next models are already being designed by AI.
Regulators are trying to get closer to frontier









