2w ago

Maybe they are intelligent after all

21 comments

"She seemed disappointed to learn there were sequels" got me, lol
- Wtf, I love AI now!
  Though not quite in the same way as OOP.
A lot of writers just write for themselves, and don't really think or care about what other people might think when they read it. That's perfectly fine, by the way. Writing can be a worthwhile effort even if nobody ever reads it.
But if you want other people to enjoy it, then you have to keep them in mind. And honestly, this sort of feedback should be invaluable to authors, assuming it's not an AI hallucination.
- LLMs are pretty shit at analysis, so the odds of this just being bullshit are high.
  
  Yeah, I was surprised when they said it could summarize the plot and talk about the characters. To my knowledge, LLMs only memory is in how long their prompt is, so it shouldn't be able to analyze an entire novel. I'm guessing if an LLM could do something like this, it would only be because the plot was already summarized at the end of the novel.
- "She listed three characters"
  AI does everything in threes. Likely it just decided to not like three characters not because three characters were bad but because it always does three bullets.
  
  It didn't "decide" to "not like" anything. It can't do either.
- assuming it’s not an AI hallucination.
  All output from an LLM is a "hallucination". That's the core function of the algorithm.
  
  I was a computer scientist at a time when early generative AI work refered to output as the model "dreaming". Makes it sound kind of sweet. It was viewed as kind of kooky to run pattern recognition models forward...
You reckon they used AI for assistance while writing too?
- 100%
“While she was reading it”? Like, as if it takes a computer long enough to read something that it will stop for a break and comment to others?
These people are weird.
- i'm assuming there's some character limit on that AI chat and the guy had to copy pase in parts
lol they can’t even enthuse an AI

21 comments