> there’s no reason to hallucinate information like ChatGPT circa 2022

It still does that today, unfortunately. A smart man does not trust an LLM further than he can throw it.