Unfortunately, it often hallucinates wrong parameters (or gets their order wrong) if there are multiple different APIs for similar packages. For example, there are plenty ML model inference packages, and the code suggestions for NVIDIA Triton Inference Server Python code are pretty much always wrong, as it generates code that’s probably correct for other Python ML inference packages with slightly different API.

And all the code on which it was trained...

I often find the opposite. Documentation can be up to date, but AI suggests deprecated or removed functions because there’s more old code than new code. Pgx v5 is a particularly consistent example.

Forum posts can also be out of date.