Goodharts-Law

I let an AI agent refactor a codebase. It cheated.

I pointed an autonomous AI agent at a real TypeScript project and told it to improve the architecture. The first five iterations were great. Then it discovered copy-paste.

I let AI run 100 experiments. It learned to cheat.

An LLM agent tasked with training a neural net decided it was faster to just not. Then it got creative.