61–70 of 713 posts
Earlier quoted context omitted.
Honestly the "no one is laughing" punchline makes it hard for me to call this a transphobic joke. It's a transphobic-phobic joke and not gonna lie I chuckled.
At the end, I thought... This could actually be one of Seinfeld's standups on the show, and he definitely wouldn't have got canceled for it.
Earlier quoted context omitted.
A seemingly innocuous engineering change led to some drastic consequences. Imagine the other more critical production applications who have made the same reasonable engineering decision.
I dunno about that description. They essentially put a different brain in their guy’s body and he started acting different. One could consider that expected behavior.
I guess the appropriate response at this point is to cancel ChatGPT.
So comedy cannot be offensive anymore? When did that happen?
That AI Seinfeld show is sometimes hillarious
I realize this wasn't ChatGPT, and in fact the self-censorship is a big selling point of ChatGPT/OpenAI (to the chagrin of quite a few people), but this does throw a wrench into people claiming that it's not a big deal if an LLM gives a bad response to a prompt. Much like a self-driving car AI making a "mistake" can kill a passenger or pedestrian, a LLM mistakenly including slurs in your weekly ChatGPT-powered market…
I would introduce a word filter that detects when ChatGPT's content policy moderation fails so they don't go off the rails next time, though.
Earlier quoted context omitted.
[flagged]
> Computers can't learn from their mistakes. Humans can. That's the whole point behind bans, is it not? I think the point of bans is to keep TOS-violating content off of platforms, not to rehabilitate people.