@dave
> Anwsers judged by other doctors.
And... I lost interest. That is not a diagnosis test, but a popularity contest.
@LukeAlmighty No, the % correct answers was lower with older LLMs for a reason: because the answers they were giving were wrong. Parity with doctors indicates that LLMs have reached the point where they appear to possess the same level of expertise as your average doctor.
@dave
I know...
I was saying that already. ![]()
@LukeAlmighty You called it a popularity contest, dude
@dave
yes, I did