[CamperBob2]

But you can’t prompt your way out of hallucinations and other AI artifacts

That's not the case, and hasn't been for some time, but it sounds like your mind's made up.

[jamiek88]

Hallucinations have been solved?! That’s great news! Must have missed that.

[CamperBob2]

Hallucinations have been solved?!

Apparently not, because no one but you implied that they had been.

There are prompting strategies that improve the odds greatly, but like the GGP, you've made up your mind, so it's a waste of time to argue otherwise.

[andrekandre]

i think they are referring to statements that they have "solved" hallucinations and it wont be a problem anymore (which it obviously isn't yet anyways)

[1] https://news.ycombinator.com/item?id=44779198

[runarberg]

My guess is that post-training has gotten a lot better in the last couple of years and what people are attributing to better models are actually just traditional (non-LLM) models they place on top of the LLM which makes it appears that the model has increased in quality (including by seemingly fewer hallucination).

If this is the case it would be observed with different prompting strategies, when you find a prompt which puts more weight on the post-training models.