AgisFeb 10, 2023
Christopher Mims

amazing on so many levels:

1. naivete of the researcher who published this paper

2. that the first commenter on hacker news correctly identified what's wrong with the reasoning behind the paper

3. that the rest of the thread on HN is just nerds who don't have the foggiest notion of what a 'mind' is arguing whether a stochastic parrot that has memorized and regurgitates texts written by beings with one, itself has a mind

https://news.ycombinator.com/item?id=34730365

Theory of Mind May Have Spontaneously Emerged in Large Language Models | Hacker News

Feb 9, 2023 at 11:03pmWeb
Show threadVitor PyFeb 10, 2023
@mimsical Point 3 is not really amazing, that's pretty much what I expect from most HN comments.
Show threadMortenFeb 10, 2023
@mimsical @Riedl
Oh God, it's the asshole (from Stanford) who tried to build an AI for detecting people's sexual orientation from images.
Show threadBrad WybleFeb 10, 2023

@mimsical lots of wildly unsubstantiated facts being reported here:

E.g.

"Some interesting facts that point to it being a difference of degree. LLM are actually are more accurate when asked to explain their thinking. They make similar mistakes to humans intuitive reasoning."

Show threadChristopher MimsFeb 10, 2023
@bwyble ...yeah
Show threadLara J. MartinFeb 10, 2023
@mimsical Did a LLM write this?
Show threadChristopher MimsFeb 12, 2023
@laramar honestly maybe yes?