“misrepresent” is a vague term. Actual graph from the study
The main issue is usual… sources. AI is bad at sources without a proper pipeline. They note that Gemini is the worst at 72%.
Note, they’re not testing models with their own pipeline. They’re testing other people’s products. This is more indicative of the product design than the actual models
I’ve had someone else’s AI summarize some content I created elsewhere, and it got it incredibly wrong to the point of changing the entire meaning of my original content.
Makes sense. I have used AI for software development tasks such as manipulating SQL queries and XML files (tedious things) and am always disappointed with how AI will misinterpret some things. But it’s obvious with those when the requests fail. But for things like “the news” where there is no QA team to point out the defect, it will be much harder to notice. And when AI starts (or continues) to use AI generated posts as sources, it will get much worse.
I dont understand the use people make of AI. I know a lot of of professionnal composer who are like “That’s awesome, AI does the music for me now!” and I’m like, cool, now you only have the boring part of the job to do since the fun part was made by AI. Creating the music is litteraly the only fun part, I hate everything around it.
It’s a word predictor. It is good at simple text processing. Think local code refactoring, changing the style or structure of a small text piece, or summarizing small text pieces into even smaller text pieces. It is ok at synthesizing new text that has similar structure to the training corpus. Think generating repetitive boilerplate or copywriting. It is very bad at recalling or checking facts, logic, mathematics, and everything else that people seem to be using it for nowadays.
The AI creating music is not an LLM
Ah, sorry, missed the context
There’s a few replies talking about humans misrepresenting the news. This is true, but part of the problem here is that most people understand the concept of bias - even if only to the extent of “my people neutral, your people biased”. But this is less true for LLMs. There’s research which shows that because LLMs present information authoritatively that not only do people tend to trust them, but they’re actually less likely to check the sources that the LLM provides than they would be with other forms of being presented with information.
And it’s not just news. I’ve seen people seriously argue that fringe pseudo-science is correct because they fed a very leading prompt into a chatbot and got exactly the answer they were looking for.
I wonder if people trust ChatGPT more or less than an international celebrity who is also their best friend.
The info-sphere today is already a highly delusional place and news can be often contradictory, even from day to day especially by outlets like BCC who is more focused on setting global narratives, not being a reporter of facts as best understood at the moment. No wonder AI would be confused, most readers are confused when navigating every statement made by experts or anonymous officials on every subject. Seems like this study really measured an AI models ability to vomit out the same text in different words and avoiding using any outside context be it accurate or hallucination.
BCC
Yet the LLM seems to be what everyone is pushing, because it will supposedly get better. Haven’t we reached the limits of this model and shouldn’t other types of engines be tried?
shouldn’t other types of engines be tried?
Sure, but the tricky bit is to be more specific than that.
Well, you know…
“Waves vaguely”
I wish they had broke it out by AI. The article states:
“Gemini performed worst with significant issues in 76% of responses, more than double the other assistants, largely due to its poor sourcing performance.”
But I don’t see that anywhere in the linked PDF of the “full results”.
This sort of study should also be re-done from time to time to track AI version numbers.
It doesn’t really matter, “AI” is being asked to do a task it was never meant to do. It isn’t good at it, and it will never be good at it.
Using an LLM to return accurate information is like using a shoe to hammer a nail.
We’ve all done it?
Nope, my soles are too soft.
Except that a shoe is vaguely hammer ish. More like pounding a screw in with your forehead.
And then I wonder how frequently humans misinterpret the mistranslated news.
Humans do it often, but they don’t have billions of dollars funding their responses.
Worse: One third of adult actually believe the shit the AI produces.
Will they change their disclaimer now, from “can be wrong” to “is often wrong”? /s
Parrot is wrong almost half of the time. Who knew?
Do you realize what you just said!!!
Wow! They have reached parrot intelligence!
Next they might teach it to butterfly! You know, like you’re off the ground and going somewhere in open air, but they just keep building shit right where you’re flying… And lamps!
From there, who knows?!
deleted by creator
The fascist social media influencers are already pushing generated bodycam and surveillance videos for xenophobia etc. A large enough mass of the population doesn’t know what’s real and that’s the goal
deleted by creator
this will sound far reaching, but what would you think about holding internet literacy talks with your (chronological age) peers?
Precision, nuance, and up to the moment contextual understanding are all missing from the “intelligence.”
Like the average American with an 8th grade reading comprehension.
Which is what they used for the training data.
So it’s about on par with humans, then.
buT AI iS hERe tO StAY
So less of a percentage than the readers and mass media
