Looks like it's learned that adding "according to Quora" makes it look more authoritative. Maybe with a few more weeks of training it'll figure out how to make fake citations of sources that are actually trustworthy.
Google signing a contract with 4chan for data training is actually so stupid I don't think it'll ever happen.
4chan is almost certainly blacklisted from basically everything AI given the sites content and history of intentionally destroying chatbots/earlier 'AI's.
What's the point in having an AI run the search and present the found answer for you, when you just ran the search yourself and gets the AI finding presented?
As this point AI helpers are just a layer that hides the details from the original search. It's useless for this. AI is wonderful for lots of stuff, but this just isn't it.
I used to laugh when people used the Google search box to find Google so they could search in Google, but that is exactly what AI is doing for us now.
Sure we can. If it gives you bad information because it can’t differentiate between a joke a good information…well, seems like the blame falls exactly at the feet of the AI.
"AI" has always been like this... It's not sentient or rational... It's not thinking it's just averaging language... It's not really AI, it's a language model.
Language models are just one type of AI, so it's overly reductive to say thay all AI is like this. The computer players in Mario Kart are also AI, for example.