SEMANTIC CLOSURE: WHY COMPILERS KNOW WHEN THEY ARE RIGHT AND LLMS DO NOT

Summary from the TLDR AI newsletter

Compilers are deterministic, and large language models are stochastic.