503
We have to stop ignoring AI’s hallucination problem
(www.theverge.com)
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 17 May 2024
503 points (94.8% liked)
Technology
57273 readers
4528 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
I fucking hate how OpenAi and other such companies claim their models "understand" language or are "fluent" in French. These are human attributes. Unless they made a synthetic brain, they can take these claims and shove them up their square tight corporate behinds.
I though I would have an aneurism reading their presentation page on Sora.
They are saying Sora can understand and simulate complex physics in 3D space to render a video.
How can such bullshit go unchallenged. It drives me crazy.
This is circular logic: only humans can be fluent, so the models can't be fluent because they aren't human.
And it's universally upvoted...in response to an ais getting things wrong so they can't be doing anything but hallucinating.
And will you learn from this? Nope. I'll just be down voted and shouted at.
It's not circular. LLMs cannot be fluent because fluency comes from an understanding of the language. An LLM is incapable of understanding so it is incapable of being fluent. It may be able to mimic it but that is a different thing. (In my opinion)
You might agree with the conclusion, and the conclusion might even be correct, but the poster effectively argued 'only humans can be fluent, and it's not a human so it isn't fluent' and that is absolutely circular logic.
If we ignore the other poster, do you think the logic in my previous comment is circular?
Hard to say. You claim they are incapable of understanding, which is why they can't be fluent. however, really, the whole argument boils down to whether they are capable of understanding. You just state that as if it's established fact, and I believe that's an open question at this point.
So whether it is circular depends on why you think they are incapable of understanding. If it's like the other poster, and it's because that's a human(ish) only trait, and they aren't human...then yes.
This is not at all what I said. If a machine was complex enough to reason, all power to it. But these LLMs cannot.