Isnt it already in AI as the prior version were publicly and should be in training corpus?

The World Factbook was updated weekly. This was because facts changed.

Every time an article like this gets posted some commenter INEVITABLY brings up "isn't this solved because AI" and god it is so depressing. Apparently a whole lot of people out there existing in the world genuinely think fucking LLMs are going to be reliable stewards of knowledge.

We are fucking cooked.

Ai training can be thought of like human training (school), much of what you learn shapes you but you forget the details. We need to continue to have real sources of info.

Sure but that doesn't mean it'll perfectly retrieve information it's trained on. There's a lot of conflicting sources, hallucinations, etc.