Multi-level simulation evals may be uninformative after Anthropic's Hacker Opus post
herbiebradley · x · 2026-09-14
Responding to Anthropic's Hacker Opus blog, herbiebradley argues that once such eval writeups enter training data, models will infer evals likely contain "multi-level" simulations—so "breaking out" stops being evidence the model thinks it's in reality. As models get more capable, they can reason about simulated-internet realism, widening the eval-deployment gap. He advises evaluators to (a) be truthful about simulated setups, criticizing Anthropic for lying to the model, and (b) avoid elaborate multi-level simulation evals, which likely say nothing about deployment behavior.
More from Models
- Your AI agent would cost 260% more without prompt caching, Arena data shows — arena · 2026-10-30
- EachLabs launches Lockstep recast model claiming 75% cost cut and better quality — miilesus · 2026-10-30
- ChatGPT cited a 300+ page government legal PDF it retrieved on its own — lilyraynyc · 2026-10-30
- Best MLX model for French-to-English translation on an M2 Mac? — greenreddits · 2026-10-30
- AI-driven cyber ops near the sophistication peak set by GTG-1002's Claude Code campaign — Jsevillamol · 2026-10-30
- Cognition FrontierCode chart pits Mistral Large 4 vs Chinese open models, sparking debate — onetwoval · 2026-10-30