I Reverse-Engineered ChatGPT's Retrieval Stack. The Bottleneck Isn't What You Think.

ChatGPT cites its sources. You see the neat little [1], [2] markers, and the implicit message is: the model went out, looked at the web, brought back evidence, and is showing you receipts. That story is half right. The other half is what every team building a RAG system gets wrong. There is no single retrieval system inside ChatGPT. There are at least two — a parametric one frozen in the weights and a live one that fires only sometimes plus a tool layer deciding which to invoke, plus a generat...