Submitted 4 weeks ago by will_a113@lemmy.ml to technology@lemmy.world
Using Reddit’s popular ChangeMyView community as a source of baseline data, OpenAI had previously found that 2022’s ChatGPT-3.5 was significantly less persuasive than random humans, ranking in just the 38th percentile on this measure. But that performance jumped to the 77th percentile with September’s release of the o1-mini reasoning model and up to percentiles in the high 80s for the full-fledged o1 model.
So are you smarter than a Redditor?
If you don’t read the article, this sounds worse than it is. I think this is the important part:
ChatGPT’s persuasion performance is still short of the 95th percentile that OpenAI would consider “clear superhuman performance,” a term that conjures up images of an ultra-persuasive AI convincing a military general to launch nuclear weapons or something. It’s important to remember, though, that this evaluation is all relative to a random response from among the hundreds of thousands posted by everyday Redditors using the ChangeMyView subreddit. If that random Redditor’s response ranked as a “1” and the AI’s response ranked as a “2,” that would be considered a success for the AI, even though neither response was all that persuasive.
OpenAI’s current persuasion test fails to measure how often human readers were actually spurred to change their minds by a ChatGPT-written argument, a high bar that might actually merit the “superhuman” adjective. It also fails to measure whether even the most effective AI-written arguments are persuading users to abandon deeply held beliefs or simply changing minds regarding trivialities like whether a hot dog is a sandwich.
This is the buried lede that’s really concerning I think.
Their goal is to create AI agents that are indistinguishable from humans and capable of convincing people to hold certain positions.
Some time in the future all online discourse may be just a giant AI fueled tool sold to the highest bidders to manufacture consent.
Their goal is to create AI agents that are indistinguishable from humans and capable of convincing people to hold certain positions.
A very large portion of people, possibly more than half, do change their views to fit in with everyone else. So an army of bots pretending to have a view will sway a significant portion of the population just through repetition and exposure with the assumption that most other people think that way.
It’s no surprise that social media companies are working on AI their platforms are no longer social, they are just tools to control public opinion.
Other governments and oligarchs will pay any money to have that kind of power.
It already is, at least on Armenia and Azerbaijan.
Genius.
I mean… one one hand it’s hardly supprising. Off the bat we know AI is more knowledgable than any single individual that doesn’t bother to research… and well 80% of online forum type posts aren’t exactly researched. second, AI can confidently bullshit in a way that can only be debunked easily by someone knowledgeable.
Hmmm, you’re pretty convincing and well researched for a human…
OpenAI: our models are 82% of Redditors
Do they think that’s a high bar?lol
So open ai is admitting to botting comments on reddit. To be honest with how shit reddit is I actually rather read ai comments than the same stupid reddit meme being repeated for the last decade.
Bot on bot crime.
Aside from a shinking number of subs the only thing redditors can convince me is that I should stop looking at reddit. So if that’s your bar . . .
Lets put it the other way round: there are a lot of people in social networks who are dumb enough that them being overtaken by an AI is no real surprise.
No need to limit it to only people on social media…
Their models are more persuasive than a person and/or older model with internet access. Very impressive. I wager your stock is worth all of the gold in fort knox ($0).
Their own older model, no less.
It would be weirder/more of note if their new model was worse.
Comparing Assumed Intelligence with an average Redditor is like asking: Are you smarter than a fifth grader?
Hint: Nope.
You mean those aggressive morons more Redditish than you expect from even Reddit are bots? Not a surprise.
rtxn@lemmy.world 4 weeks ago
That bar is so low it’s practically a tripping hazard in hell.
will_a113@lemmy.ml 4 weeks ago
😂