This is a really interesting paragraph to me because I definitely think these results shouldn’t be published or we’ll only get more of these “whoopsie” experiments.
At the same time though, I think it is desperately important to research the ability of LLMs to persuade people sooner rather than later when they become even more persuasive and natural-sounding. The article mentions that in studies humans already have trouble telling the difference between AI written sentences and human ones.
This is certainly not the first time this has happened. There’s nothing to stop people from asking ChatGPT et al to help them argue. I’ve done it myself, not letting it argue for me but rather asking it to find holes in my reasoning and that of my opponent. I never just pasted what it said.
I also had a guy post a ChatGPT response at me (he said that’s what it was) and although it had little to do with the point I was making, I reasoned that people must surely be doing this thousands of times a day and just not saying it’s AI.
To say nothing of state actors, “think tanks,” influence-for-hire operations, etc.
The description of the research in the article already conveys enough to replicate the experiment, at least approximately. Can anyone doubt this is commonplace, or that it has been for the last year or so?
black on white, ew
I’m pretty sure that only applies due to a majority of people being morons. There’s a vast gap between the 2% most intelligent, 1/50, and the average intelligence.
Also please put digital text on white on black instead of the other way around
I agree, but that doesn’t change anything, right? Even if you are in the 2% most intelligent and you’re somehow immune, you still have to live with the rest who do get influenced by AI. And they vote. So it’s never just a they problem.
What? Intelligent people get fooled all the time. The NXIVM cult was made up mostly of reasonably intelligent women. Shit that motherfucker selected for intelligent women.
You’re not immune. Even if you were, you’re incredibly dependent on people of average to lower intelligence on a daily basis. Our planet runs on the average intelligence.
This research is good, valuable and desperately needed. The uproar online is predictable and could possibly help bring attention to the issue of LLM-enabled bots manipulating social media.
This research isn’t what you should get mad it. It’s pretty common knowledge online that Reddit is dominated by bots. Advertising bots, scam bots, political bots, etc.
Intelligence services of nation states and political actors seeking power are all running these kind of influence operations on social media, using bot posters to dominate the conversations about the topics that they want. This is pretty common knowledge in social media spaces. Go to any politically charged topic on international affairs and you will notice that something seems off, it’s hard to say exactly what it is… but if you’ve been active online for a long time you can recognize that something seems wrong.
We’ve seen how effective this manipulation is on changing the public view (see: Cambridge Analytica, or if you don’t know what that is watch ‘The Great Hack’ documentary) and so it is only natural to wonder how much more effective online manipulation is now that bad actors can use LLMs.
This study is by a group of scientists who are trying to figure that out. The only difference is that they’re publishing their findings in order to inform the public. Whereas Russia isn’t doing us the same favors.
Naturally, it is in the interest of everyone using LLMs to manipulate the online conversation that this kind of research is never done. Having this information public could lead to reforms, regulations and effective counter strategies. It is no surprise that you see a bunch of social media ‘users’ creating a huge uproar.
Most of you, who don’t work in tech spaces, may not understand just how easy and cheap it is to set something like this up. For a few million dollars and a small staff you could essentially dominate a large multi-million subscriber subreddit with whatever opinion you wanted to push. Bots generate variations of the opinion that you want to push, the bot accounts (guided by humans) downvote everyone else out of the conversation and, in addition, moderation power can be seized, stolen or bought to further control the conversation.
Or, wholly fabricated subreddits can be created. A few months prior to the US election there were several new subreddits which were created and catapulted to popularity despite just being a bunch of bots reposting news. Now those subreddits are high in the /all and /popular feeds, despite their moderators and a huge portion of the users being bots.
We desperately need this kind of study to keep from drowning in a sea of fake people who will tirelessly work to convince you of all manner of nonsense.
Regardless of any value you might see from the research, it was not conducted ethically. Allowing unethical research to be published encourages further unethical research.
This flat out should not have passed review. There should be consequences.
If the need was justified big enough and negative impact low enough, it could pass review. The lack of informed consent can be justified with sufficient need and if consent would impact the science. The burden is high but not impossible to overcome. This is an area with huge societal impact so I would consider an ethical case to be plausible.
Conversely, while the research is good in theory, the data isn’t that reliable.
The subreddit has rules requiring users engage with everything as though it was written by real people in good faith. Users aren’t likely to point out a bot when the rules explicitly prevent them from doing that.
There wasn’t much of a good control either. The researchers were comparing themselves to the bots, so it could easily be that they themselves were less convincing, since they were acting outside of their area of expertise.
And that’s even before the whole ethical mess that is experimenting on people without their consent. Post-hoc consent is not informed consent, and that is the crux of human experimentation.
Users aren’t likely to point out a bot when the rules explicitly prevent them from doing that.
In fact one user commented that he had his comment calling out one of the bots as a bot deleted by mods for breaking that rule
as opposed to thousands of bots used by russia everyday on politics related subs.
On all subs.
There’s no guarantee anyone on there (or here) is a real person or genuine. I’ll bet this experiment has been conducted a dozen times or more but without the reveal at the end.
I’ve worked in quite a few DARPA projects and I can almost 100% guarantee you are correct.
Some of us have known the internet has been dead since 2014
Hello, this is John Cleese. If you doubt that this is the real John Cleese, here is my mother to confirm that I am, in fact, me. Mother! Am I me?
Oh yes!
There you have it. I am me.
Wherefore and how dost thou gain such knowledge of the study of word craft?
Shall we talk about Eglin Airforce base or Jessica Ashoosh?
Shall I compare thee to a Summer’s day?
If you think that, the US is the only country that does this. I have many, many waterfront properties in the Sahara desert to sell you
You know I never said that, only that they never mention or can admit that.
The american bots or online operatives always need to start crying about Russian or Chinese interference on any unrelated subject?
Like this Shakleford here, who admits he’s worked for the fascist imperialist warcriminal state.
I’ve seen plenty of US bootlicker bots/operatives and hasbara genocider scum. I can smell them from far.
Not so much Chinese or Russians.Well my friend, if you can’t smell the shit you should probably move away from the farm. Russian and Chinese has a certain scent to it. The same with American. Sounds like you’re just nose blind.
I know anything said online that goes against the western narrative immediately gets slandered: ‘Russian bots’, ‘100+ social credit’ and that lame BS.
Paranoid delusional Pavlovian reflexes induced by western propaganda.
Incapable of fathoming people have another opinion, they must be paid!
If that’s the mindset hen you will see indeed a lot of those.
The most obvious ones to spot are definitely the Hasbara types, same pattern and vocab, and really bad at what they do.I mean that’s just like your opinion man.
However, there are for a fact government assets promoting those opinions and herding those clueless people. What a lot of people failed to realize is that this isn’t a 2v1 or even a 3v1 fight. This is an international free-for-all with upwards of 45 different countries getting in on the melee.
Dozens? That’s like saying there are hundreds of ants on earth. I’m very comfortable saying it’s hundreds, thousands, tens of thousands. And I wouldn’t be surprised if it’s hundreds of thousands of times.
There’s no guarantee anyone on there (or here) is a real person or genuine.
I’m pretty sure this isn’t a baked-in feature of meatspace either. I’m a fan of solipsism and Last Thursdayism personally. Also propaganda posters.
The CMV sub reeked of bot/troll/farmer activity, much like the amitheasshole threads. I guess it can be tough to recognize if you weren’t there to see the transition from authentic posting to justice/rage bait.
We’re still in the uncanny valley, but it seems that we’re climbing out of it. I’m already being ‘tricked’ left and right by near perfect voice ai and tinkered with image gen. What happens when robots pass the imitation game?
I think the reddit user base is shifting too. It’s less “just the nerds” than it used to be. The same thing happened to Facebook. It fundamentally changed when everyone’s mom joined…
I’m sorry but as a language model trained by OpenAI, I feel very relevant to interact - on Lemmy - with other very real human beings
AI is a fucking curse upon humanity. The tiny morsels of good it can do is FAR outweighed by the destruction it causes. Fuck anyone involved with perpetuating this nightmare.
Damn this AI, posting and doing all this mayhem all by itself on poor unsuspecting humans…
Yes. Fuck the owners and fuck their machine guns.
Todays “AI” is just machine learning code. It’s been around for decades and does a lot of good. It’s most often used for predictive analytics and used to facilitate patient flow in healthcare and understand volumes of data fast to provide assistance to providers, case manager, and social workers. Also used in other industries that receive little attention.
Even some language learning machines can do good, it’s the shitty people that use it for shitty purposes that ruin it.
Sure I know what it is and what it is good for, I just don’t think the juice is worth the squeeze. The companies developing AI HAVE to shove it everywhere to make it feasible, and the doing of that is destructive to our entire civilization. The theft of folks’ work, the scamming, the deep fakes, the social media propaganda bots, the climate raping energy consumption, the loss of skill and knowledge, the enshittification of writing and the arts, the list goes on and on. It’s a deadend that humanity will regret pursuing if we survive this century. The fact that we get a paltry handful of positives is cold comfort for our ruin.
The fact that we get a paltry handful of positives is cold comfort for our ruin.
This statement tells me you don’t understand how many industries are using machine learning and how many lives it saves.
That’s great. We can schedule it like heroin for professional use only, then.
I disagree. It may seem that way if that’s all you look at and/or you buy the BS coming from the LLM hype machine, but IMO it’s really no different than the leap to the internet or search engines. Yes, we open ourselves up to a ton of misinformation, shifting job market etc, but we also get a suite of interesting tools that’ll shake themselves out over the coming years to help improve productivity.
It’s a big change, for sure, but it’s one we’ll navigate, probably in similar ways that we’ve navigated other challenges, like scams involving spoofed webpages or fake calls. We’ll figure out who to trust and how to verify that we’re getting the right info from them.
LLMs are not like the birth of the internet. LLMs are more like what came after when marketing took over the roadmap. We had AI before LLMs, and it delivered high quality search results. Now we have search powered by LLMs and the quality is dramatically lower.
Sure, and we had an internet before the world wide web (ARPANET). But that wasn’t hugely influential until it was expanded into what’s now the Internet. And that evolved into the world wide web after 20-ish years. Each step was a pretty monumental change, and built on concepts from before.
LLMs are no different. Yes they’re built on older tech, but that doesn’t change the fact that they’re a monumental shift from what we had before.
Let’s look at access to information and misinformation. The process was something like this:
- Physical encyclopedias, newspapers, etc
- Digital, offline encyclopedias and physical newspapers
- Online encyclopedias and news
- SEO and the rise of blog/news spam - misinformation is intentional or negligent
- Early AI tools - misinformation from hallucinations is largely also accidental
- Misinformation in AI tools becomes intentional
We’re in the transition from 5 to 6, which is similar to the transition from 3 to 4. I’m old enough to have seen each of these transitions.
The way people interact with the world is fundamentally different now than it was before LLMs came out, just like the transition from offline to online computing. And just like people navigated the transition to SEO nonsense, people need to navigate he transition to LLM nonsense. It’s quite literally a paradigm shift.
Enshittification is a paradigm shift, but not one we associate with the birth of the internet.
On to your list. Why does misinformation appear after the birth of the internet? Was yellow journalism just a historical outlier?
What you’re witnessing is the “Red Queen hypothesis”. LLMs have revolutionized the scam industry and step 7 is an AI arms race against and with misinformation.
Why does misinformation appear after the birth of the internet?
It certainly existed before. Physical encyclopedias and newspapers weren’t perfect, as they frequently followed the propaganda line.
My point is that a lot of people seem to assume that “the internet” is somewhat trustworthy, which is a bit bizarre. I guess there’s the fallacy that if something is untrustworthy, it won’t get attention, but instead things are given attention if they’re popular, by some definition of “popular” (i.e. what a lot of users want to see, what the platform wants users to see, etc).
Red Queen hypothesis
Well yeah, every technological innovation will be used for good and ill. The Internet gave a lot of people a voice who didn’t have it before, and sometimes that was good (really helpful communities) and sometimes that was bad (scam sites, misinformation, etc).
My point is that AI is a massive step. It can massively increase certain types of productivity, and it can also massively increase the effectiveness of scams and misinformation. Whichever way you look at it, it’s immensely impactful.
Personally I love how they found the AI could be very persuasive by lying.
why wouldn’t that be the case, all the most persuasive humans are liars too. fantasy sells better than the truth.
I mean, the joke is that AI doesn’t tell you things that are meaningfully true, but rather is a machine for guessing next words to a standard of utility. And yes, lying is a good way to arbitrarily persuade people, especially if you’re unmoored to any social relation with them.
If anyone wants to know what subreddit, it’s r/changemyview. I remember seeing a ton of similar posts about controversial opinions and even now people are questioning Am I Overreacting and AITAH a lot. AI posts in those kind of subs are seemingly pretty frequent. I’m not surprised to see it was part of a fucking experiment.
AI posts or just creative writing assignments.
Right. Subs like these are great fodder for people who just like to make shit up.
lmao wait what holy shit is that line originally from Arthur??? that sounds exactly like something the bunny dude would say
This was comments, not posts. They were using a model to approximate the demographics of a poster, then using an LLM to generate a response counter to the posted view tailored to the demographics of the poster.
You’re right about this study. But, this research group isn’t the only one using LLMs to generate content on social media.
There are 100% posts that are bot created. Do you ever notice how, on places like Am I Overreacting or Am I the Asshole that a lot of the posts just so happen to hit all of the hot button issues all at once? Nobody’s life is that cliche, but it makes excellent engagement bait and the comment chain provides a huge amount of training data as the users argue over the various topics.
I use a local LLM, that I’ve fine tuned, to generate replies to people, who are obviously arguing in bad faith, in order to string them along and waste their time. It’s setup to lead the conversation, via red herrings and other various fallacies to the topic of good faith arguments and how people should behave in online spaces. It does this while picking out pieces of the conversation (and from the user’s profile) in order to chastise the person for their bad behavior. It would be trivial to change the prompt chains to push a political opinion rather than to just waste a person/bot’s time.
This is being done as a side project, on under $2,000 worth of consumer hardware, by a barely competent progammer with no training in Psychology or propaganda. It’s terrifying to think of what you can do with a lot of resources and experts working full-time.
You know what Pac stands for? PAC. Program and Control. He’s Program and Control Man. The whole thing’s a metaphor. All he can do is consume. He’s pursued by demons that are probably just in his own head. And even if he does manage to escape by slipping out one side of the maze, what happens? He comes right back in the other side. People think it’s a happy game. It’s not a happy game. It’s a fucking nightmare world. And the worst thing is? It’s real and we live in it.
Please elaborate. I would love to understand this from black mirror but I don’t get it.
Lol, coming from the people who sold all of your data with no consent for AI research
The quote is not coming from Reddit, but from a professor at Georgia Institute of Technology
The reason this is “The Worst Internet-Research Ethics Violation” is because it has exposed what Cambridge Analytica’s successors already realized and are actively exploiting. Just a few months ago it was literally Meta itself running AI accounts trying to pass off as normal users, and not an f-ing peep - why do people think they, the ones who enabled Cambridge Analytica, were trying this shit to begin with. The only difference now is that everyone doing it knows to do it as a “unaffiliated” anonymous third party.
One of the Twitter leaks showed a user database that effectively had more users than there were people on earth with access to the Internet.
Before Elon bought the company he was trashing them on social media for being mostly bots. He’s obviously stopped that now that he was forced to buy it, but the fact that Twitter (and, by extension, all social spaces) are mostly bots remains.
Just a few months ago it was literally Meta itself…
Well, it’s Meta. When it comes to science and academic research, they have rather strict rules and committees to ensure that an experiment is ethical.
The headline is that they advertised beauty products to girls after they detected them deleting a selfie. No ethics or morals at all
You may wish to reword. The unspecified “they” reads like you think Meta have strict ethical rules. Lol.
Meta have no ethics whatsoever, and yes I assume you meant universities have strict rules however the approval of this study marks even that as questionable
The ethics violation is definitely bad, but their results are also concerning. They claim their AI accounts were 6 times more likely to persuade people into changing their minds compared to a real life person. AI has become an overpowered tool in the hands of propagandists.
It would be naive to think this isn’t already in widespread use.
I mean that’s the point of research: to demonstrate real world problems and put it in more concrete terms so we can respond more effectively
To be fair, I do believe their research was based on how convincing it was compared to other Reddit commenters, rather than say, an actual person you’d normally see doing the work for a government propaganda arm, with the training and skillset to effectively distribute propaganda.
Their assessment of how “convincing” it was seems to also have been based on upvotes, which if I know anything about how people use social media, and especially Reddit, are often given when a comment is only slightly read through, and people are often scrolling past without having read the whole thing. The bots may not have necessarily optimized for convincing people, but rather, just making the first part of the comment feel upvote-able over others, while the latter part of the comment was mostly ignored. I’d want to see more research on this, of course, since this seems like a major flaw in how they assessed outcomes.
This, of course, doesn’t discount the fact that AI models are often much cheaper to run than the salaries of human beings.
This, of course, doesn’t discount the fact that AI models are often much cheaper to run than the salaries of human beings.
And the fact that you can generate hundreds or thousands of them at the drop of a hat to bury any social media topic in highly convincing ‘people’ so that the average reader is more than likely going to read the opinion that you’re pushing and not the opinion of the human beings.
This is probably the most ethical you’ll ever see it. There are definitely organizations committing far worse experiments.
Over the years I’ve noticed replies that are far too on the nose. Probing just the right pressure points as if they dropped exactly the right breadcrumbs for me to respond to. I’ve learned to disengage at that point. It’s either they scrolled through my profile. Or as we now know it’s a literal psy-op bot. Already in the first case it’s not worth engaging with someone more invested than I am myself.
Yeah I was thinking exactly this.
It’s easy to point to reasons why this study was unethical, but the ugly truth is that bad actors all over the world are performing trials exactly like this all the time - do we really want the only people who know how this kind of manipulation works to be state psyop agencies, SEO bros, and astroturfing agencies working for oil/arms/religion lobbyists?
Seems like it’s much better long term to have all these tricks out in the open so we know what we’re dealing with, because they’re happening whether it gets published or not.
actors all over the world are performing trials exactly like this all the time
I marketing speak this is called A/B testing.
Over the years I’ve noticed replies that are far too on the nose. Probing just the right pressure points as if they dropped exactly the right breadcrumbs for me to respond to. I’ve learned to disengage at that point. It’s either they scrolled through my profile. Or as we now know it’s a literal psy-op bot. Already in the first case it’s not worth engaging with someone more invested than I am myself.
You put it better than I could. I’ve noticed this too.
I used to just disengage. Now when I find myself talking to someone like this I use my own local LLM to generate replies just to waste their time. I’m doing this by prompting the LLM to take a chastising tone, point out their fallacies and to lecture them on good faith participation in online conversations.
It is horrifying to see how many bots you catch like this. It is certainly bots, or else there are suddenly a lot more people that will go 10-20 multi-paragraph replies deep into a conversation despite talking to something that is obviously (to a trained human) just generated comments.
Would you mind elaborating? I’m naive and don’t really know what to look for…
I think the simplest way to explain it is that the average person isn’t very skilled at rhetoric. They argue inelegantly. Over a long time of talking online, you get used to talking with people and seeing how they respond to different rhetorical strategies.
In these bot infested social spaces it seems like there are a large number of commenters who just seem to argue way too well and also deploy a huge amount of fallacies. This could be explained, individually, by a person who is simply choosing to argue in bad faith; but, in these online spaces there seem to be too many commenters who seem to deploy these tactics compared to the baseline that I’ve established in my decades of talking to people online.
In addition, what you see in some of these spaces are commenters who seem to have a very structured way of arguing. Like they’ve picked your comment apart into bullet points and then selected arguments against each point which are technically on topic but misleading in a way.
I’ll admit that this is all very subjective. It’s entirely based on my perception and noticing patterns that may or may not exist. This is exactly why we need research on the topic, like in the OP, so that we can create effective and objective metrics for tracking this.
For example, if you could somehow measure how many good faith comments vs how many fallacy-laden comments in a given community there would likely be a ratio that is normal (i.e. there are 10 people who are bad at arguing for every 1 person who is good at arguing and, of those skilled arguers 10% of them are commenting in bad faith and using fallacies) and you could compare this ratio to various online topics to discover the ones that appear to be botted.
That way you could objectively say that on the topic of Gun Control on this one specific subreddit we’re seeing an elevated ratio of bad faith:good faith scoring commenters and, therefore, we know that this topic/subreddit is being actively LLM botted. This information could be used to deploy anti-bot counter measures (captchas, for example).
Thanks for replying
Do you think response time could also indicate that a user is a bot? I’ve had an interaction that I chalked up to someone using AI, but looking back now I’m questioning if there was much human involvement at all just due to how quickly the detailed replies were coming in…
It depends, but it’d be really hard to tell. I type around 90-100 WPM, so my comment only took me a few minutes.
If they’re responding within a second or two with a giant wall of text it could be a bot, but it may just be a person who’s staring at the notification screen waiting to reply. It’s hard to say.
Reddit’s chief legal officer, Ben Lee, wrote that the company intends to “ensure that the researchers are held accountable for their misdeeds.”
What are they going to do? Ban the last humans on there having a differing opinion?
Next step for those fucks is verification that you are an AI when signing up.
Like the 90s/2000s - don’t put personal information on the internet, don’t believe a damned thing on it either.
I never liked the “don’t believe anything you read on the internet” line, it focuses too much on the internet without considering that you shouldn’t believe anything you read or hear elsewhere either, especially on divisive topics like politics.
You should evaluate information you receive from any source with critical thinking, consider how easy it is to make false claims (e.g. probably much harder for a single source if someone claims that the US president has been assassinated than if someone claims their local bus was late that one unspecified day at their unspecified location), who benefits from convincing you of the truth of a statement, is the statement consistent with other things you know about the world,…
Nice try, AI
😄
Using mainstream social media is literally agreeing to be constantly used as an advertisement optimization research subject
