I don't think it's shocking that Google knew about the thread, it's that the AI outright lied saying that the meme didn't exist and then used this post as its source for the lie when the actual post doesn't support what it said in any way. Not that we should be surprised that it's wrong, Google's AI Overviews were shown to lie millions of times an hour, but it's pretty shocking how wrong it was.
I notice... the newer models, they seem like they want to just "give up" much faster, and you have to put a bunch of extra programming around it (harness, for loops and if then's etc to force it to keep trying)
Ahh yes - AI companies do indeed want it both ways. They want to be your single source for authoritative information about every subject on earth, but also want you to know everything they tell you could possibly be wrong so they can't be held accountable for results.
Can you imagine in the 1980s the Encyclopedia Brittanica salesman trying to sell you 5 figures worth of factual knowledge, only to tell you as you're writing him a check that: oh by the way we don't actually stand behind any of the information in these books, you'll need to verify any claims yourself.
It's not the bad information that makes it a lie, but the overconfident and authoritative answers which give no indication that a significant percentage of what it spews out will be completely false that elevates it to a lie. Google could change that, but they choose to be manipulative.
We call it the gaslighting machine here. Gemini will also totally unprompted repeatedly insert 10 advertising links into non-related conversations when you are a paying user, and then claims it never happened. Someone must be pretty desperate.
