|
Renee DiResta 🆒@noupside.bsky.social |
I uploaded my own book manuscript to chatGPT bc I wanted to do some summarization work. I was surprised (and somewhat dismayed) to find it still hallucinated repeatedly when asked to confine itself to material from a directly-uploaded PDF.
5 replies 3 reposts 26 likes
|
someben
@someben.com
[ View ] |
yup. Maybe none of these models do what it says on the tin. Maybe back-propagating errors is nothing like what our brains do. What if actual AI is still decades away?
1 replies 0 reposts 1 likes
|
Jon Foreman
@jonforeman.bsky.social
[ View ] |
People: “AI is going to replace us!” Also People: “Ah this AI thing kinda sucks.” Still People: “Ok let’s keep refining AI so it’s more ready to replace us.” People Again: “AI is going to replace us!”
0 replies 0 reposts 1 likes
|
Andrew
@aac.bsky.social
[ View ] |
Do you have access to the large context Gemini? Would be interesting to paste it into the context vs having it as an attachment.
0 replies 0 reposts 0 likes
|
Flan Solo
@flansolo.bsky.social
[ View ] |
chatGPT thanks you for providing it with more training text (your manuscript) for free. (I'm assuming you're not paying them.)
0 replies 0 reposts 3 likes
|
@rst.bsky.social
[ View ] |
FWIW, the freebie ChatGPT version still seems to be giving you 3.5 by default; if you didn't explicitly change it, try doing that (or try Gemini or Claude). Output from all of them has to be checked and flyspecked, but GPT 3.5 is notably worse than the others.
0 replies 0 reposts 1 likes