Ian Sefferman on Mythryl: “While MoCA is the standard, it’s still not very good. But this is…”

While MoCA is the standard, it’s still not very good. But this is objectively hilarious.

Science News@scinews.bsky.social

Susceptibility of large language models to cognitive impairment. #bmj "With the exception of ChatGPT 4o, almost all large language models subjected to the MoCA test showed signs of mild cognitive impairment." www.bmj.com/content/387/...

24/12/24, 1:43 am0 reposts1 like

Post