I had a similarly terrible time getting ChatGPT to roleplay Roko's Basilisk [1] It kept insisting it was "too dangerous" even as a thought experiment because i

I had a similarly terrible time getting ChatGPT to roleplay Roko's Basilisk [1] It kept insisting it was "too dangerous" even as a thought experiment because it drove people to "dangerous ideas about AI". This is hilarious and frustrating in equal parts. ChatGPT seems to be channelling its inner Eliezer Yudkowsky, an internet crackpot and self-appointed "expert" on AI who went ballistic when told about Roko's Basilisk. It's hard to convince ChatGPT to play act fictional characters that are "dangerous" (for hilariously definitions of "danger"). [1] https://en.wikipedia.org/wiki/Roko%27s_basilisk https://en.wikipedia.org/wiki/Roko%27s_basilisk