2 ms·

> LLM will perform poorly We've been seeing examples of this constantly since 2022. How many more do we need?