Thank you for the link. I didn’t realise that Google had a deal with Reddit as well, which explains why it was clearly indexed from Reddit.

I agree that AI doesn’t have a clue what an accurate response is. It’s just not sentient enough to differentiate between shitposting and fact. I also totally agree that an answer given from a search result HAS to be accurate, and we’re heading down a path of a misinformation super highway if LLMs are trained on incorrect data.