Was it what, just a week ago I was being called dumb for suggesting there'd be accuracy issues with this? I mean Bing had like a whole three weeks to slap this together after OpenAI first demoed it's ability to make things up.

oh only six days ago:

https://news.ycombinator.com/item?id=34699087

> This is a commonly echoed complaint but it’s largely without merit. ChatGPT spews nonsense because it has no access to information outside of its training set.

> In the context of a search engine, single shot learning with the top search results should mitigate almost all hallucination.

hows that going?