Flutterby™! : They're not gonna get much better

Axios on Sam Altman and the GPT5 launch quotes him as saying:

"I think the models are still getting better at a rapid rate," he said. "One of the things that's interesting is the models have already saturated the chat use case. They're not gonna get much better. ... The Turing test has passed."

Which... ya know... is what a bunch of us have been predicting, but if he thinks this is "the Turing test has passed", he's...

Look, I know it's problematic to talk about levels of humanity, but one of the things that has struck me about LLMs and generative AI is how it sure seems like the proponents of these things haven't spent a lot of time hanging out with developmentally disabled adults, those who have some interests and can spout facts, but are also willing to make up whatever to please the asker. My experience of LLMs is that they're like a few such DD adults I know, I have to be very careful of how I phrase questions because I know I'm going to get "yes, and..." when the answer is more likely "no, that actually never happened".

Anyway, maybe I was wrong, and that's who Sam's baseline is?