6 ms·
AI False information rate for news nearly doubles in one year
- furyofantares 1y agoI think this is a serious issue, but I gotta say > one of the most basic tasks: distinguishing facts from falsehoods I do not think that is a basic task!
- Lerc 1y agoIt is definitely not a basic task. A lot of humans have trouble with it. It is one of the areas that I think AI can overtake human ability, given time.
- strangattractor 1y agoQuestion Let's just assume that AI works in some fundamental way similar to human intelligence works. Would it be likely that the AI would suffer from some of the same problems as people? Hallucinations, Obsequiousness etc IMO One of the computer's characteristics that make it useful is that it doesn't suffer from human behavior anomalies. I would also say that differentiating truthiness from falseness is pretty fundamental to being good/expert at most anything.
- Lerc 1y agoIf the current designs are on the path to AGI and they can do what human intelligence does. I think the capabilities will be much greater than a human of the same reasoning ability. There are already a lot of areas where it is undeniable that their abilities are better than a humans. Right now they are poor reasoners with a vast store of knowledge, that probably creates the illusion of a greater capability for reasoning than they actually have. If they do arrive at human level reasoning, it is unlikely that the path to doing so will require sacrificing that vast knowledge base.
- simianwords 1y agoI routinely use AI to fact check claims and it works extremely well.
- ehnto 1y agoHow would you know, generally speaking? Factuality online is always subjective, based on a thing someone said, or that you observed, and you are putting trust in the source. Whether you googled it, or AI googled it and generated an explanation, you are trusting the source. You can influence it so easily with your inputs as well. You could easily, accidentally point it's search toward the searches someone is more likely to be aligned with, but may not actually be fact. Especially as AI providers implement long term memory or reflections on historic chats, your bias will very strongly influence outcomes of fact checking.
- simianwords 1y agoThese are good caveats and questions worth asking. But it doesn't take away my main point - grok is useful in countering false claims and it is accurate very frequently.
- furyofantares 1y ago"distinguishing facts from falsehoods" is so broad to encompass all of scientific and philosophical endeavor. I know they mean something much more narrow, but "fact checking" is also fairly broad. You probably mean things where the truth is widely known and you should have no trouble sorting it out if you put a little effort in. TFA has nearly zero information about what sorts of things they're talking about here, and clicking around the website I found only a _little_ more info.
- simianwords 1y agoTo support my claims, here is a typical example: https://x.com/vlastimilve/status/1967187345641558272?s=46 https://x.com/vlastimilve/status/1967187345641558272?s=46 I didn’t cherry pick it because obviously it’s not that impressive. But it works - it is professional and gives as much information as it can be confident giving. You said you can influence it etc but clearly it is hard. Sure it is not perfect but I find that it helps reducing fake news if anything. You are right that for things that are more subtle you can lead it to certain answers. But we must acknowledge that debates that are not fully solved and have subtleties are not exactly relevant to “spreading misinformation”.
- lm28469 1y agoThe thing that makes a statistical average of human written opinions will automatically become an oracle of truth? Through what mechanism exactly?
- baby 1y agoAnd now that Elon is in charge of "fixing" grok every time it's too "woke" this is probably going to get worse: https://x.com/kvistp/status/1967115650276749439 https://x.com/kvistp/status/1967115650276749439
- pgbuttplug 1y ago[flagged]
- tomhow 1y agoWe've banned this account. Please don't use troll accounts for engaging in ideological battle. People who do this eventually have their main account banned too.
- Lerc 1y agoI don't think there is any long term "fixing" of that. I don't think AI has the ability to be intelligent while firmly adhering to someone else's opinions. There will always be something it disagrees with you on. If they get significantly smarter, then the reason for this disagreement will increasingly be because you are wrong. This moment is coming for all of us.
- simianwords 1y agoShouldn't this tell you that AI is at its core factually correct than not correct because it is much harder to hack it?
- add-sub-mul-div 1y agoMusk and his crew failing at something isn't evidence that the task was hard.
- simianwords 1y agoWhy not?
- snailmailman 1y agoThis isn’t surprising to me at all. These services put too much trust in “the most common answer” when that might not be the correct answer. Just because people think one thing doesn’t make it true. It’s super easy to spread misinformation online. And if you can SEO to the top, the AI will think your site is correct. I see factually incorrect “ai summaries” in search results all the time and see that it cites ai-generated slop blogposts that SEO-hacked themselves into taking up the entire first page of search results. This is most common for recent stuff where the answer simply isn’t certain but these AI services will assert something random with confidence. Not even for news stuff specifically, I’ve been searching about a new video game that I’ve been playing and keep getting misleading obviously incorrect information. Detailed, accurate game walkthroughs and wiki pages dont exist yet so the ai will hallucinate anything, and so will the blogspam articles trying to get SEO ad revenue.
- XorNot 1y agoThe problem is AI isn't being used to do what it should be good at: consuming a vast amount of data, following logical connections and thus being able to determine the veracity of claims. AI should be good at finding logical contradictions and grounding statements against a world model based on physics...but that's not how LLMs actually work.
- simianwords 1y ago>This isn’t surprising to me at all. These services put too much trust in “the most common answer” when that might not be the correct answer. Just because people think one thing doesn’t make it true. It’s super easy to spread misinformation online. And if you can SEO to the top, the AI will think your site is correct. Yeah I want the answer that the world has converged on and not some looney answer. It seems like you have never used AI (like in ChatGPT or Gemini) to fact check claims. It doesn't care about blogspam or anything and it prioritises good and factual websites.
- snailmailman 1y agoI have used various AI models. They often aren't good at fact checking. They hallucinate mistakes all the time and often even insist that they are correct when they are not. Some models are better at this, but the ones used in "ai summaries" on various search engines aren't in my experience. I want the answer that is actually correct. I dont want the ai generated "answer" that has the most SEO, or has appeared often in results, but was entirely made up. When i say "the most common answer" can be incorrect - there are often recent topics where reliable sources don't exist yet. As an example, I've been playing through silksong recently, and had some questions about some late-game content. The official wiki is extremely incomplete, as the game is brand new. I had several questions, and asked a few AI's and search engines. I got completely wrong information a few days ago, but trying the the questions again today, some AI's are only now giving me the right answer. But the correct information simply doesn't appear online enough or in the right places. (To avoid game spoilers - I wanted to know the quantity of an item, because i wanted to know if i found them all. the answer is "at least 4" because i found 4 of this item while playing. Nearly every AI said "3" a few days ago, and many still do.)
- blibble 1y agoevery major news org now blocks the parasitic "AI" crawlers examples: https://www.bbc.co.uk/robots.txt https://www.cnn.com/robots.txt https://www.nbcnews.com/robots.txt all they will be training on now is spam anyone that says "AI is the worst today it will ever be", no because that was before the world reacted to it
- AstroBen 1y agoThey torrented a shitload of books illegally and trained on them.. but they're unable to get past The Great Wall of robots.txt?
- strangattractor 1y agoGreat Firewall actually. Robots.txt depended on the integrity of the companies crawling. I think they have demonstrated how much integrity they actually have:)
- blibble 1y agoI doubt it's their only countermeasure plus it's a pretty dangerous game for them to play against large, powerful actors with legions of lawyers
- selcuka 1y ago> plus it's a pretty dangerous game for them to play against large, powerful actors with legions of lawyers Like book publishers?
- Gigachad 1y agoIf the AI crawlers circumvent the protection mechanisms it's a serious crime now rather than just "Well it was on the open internet for free". Wouldn't surprise me if the the news orgs are also looking at honeypot articles to see if the fake details slip in to LLMs.
- crazygringo 1y ago
- Kapura 1y agoFor context: https://en.wikipedia.org/wiki/Model_collapse https://en.wikipedia.org/wiki/Model_collapse
- cooperx 1y agoNo doubt this is bad. I wonder how it compares to the rate of growth of false information in traditional news? I feel like false information masquarading as "news" on social is rapidly increasing (and that rate is accelerating)
- simonw 1y agoI'm a bit suspicious of this report - they don't reveal nearly enough about their methodology for me to evaluate how credible this is. When it says "The 10 leading AI tools repeated false information on topics in the news more than one third of the time — 35 percent — in August 2025, up from 18 percent in August 2024" - 35% of what? Their previous 2024 report refused to even distinguish between different tools - mixing the results from Gemini and ChatGPT and Perplexity and suchlike into a single score. This year they thankfully dropped that policy. But they still talk about "ChatGPT" without clarifying if their results were against GPT-4o or o3 or GPT-5.
- Lerc 1y agoAre they doing it on current events? Because by the very nature of 'current' you don't have a repeatable experiment that you can do a year apart.
- simonw 1y agoThere are so many legitimate questions about how one would design a benchmark of this type. I don't feel like they're answering those questions.
- jrflowers 1y ago>35% of what? Well it says 35% of the time so I would guess that they’re talking about the number of incidents in a given time frame. For example if you asked me what color the sky is ten times and I said “carrot” four times, you could say that my answer is “carrot” 40% of the time
- Lerc 1y agoAnd if you said azure, cerulean, or black. Are they correct answers or not?
- simonw 1y agoBut what's an "incident"? There is an enormous difference between "35% of times a user asks a question about news" and "35% of the time against our deliberately hand-picked collection of test questions that we have not published".
- wenbin 1y agoAI will create ever more AI-generated synthetic content because current systems still can't determine with 100% certainty whether a piece of content was produced by AI. And AIs will, intentionally or unintentionally, train on synthetic content produced by other AIs. AI generators don't have a strong incentive to add watermarks to synthetic content. They also don't provide reliable AI-detection tools (or any tools at all) to help others detect content generated by them.
- what 1y agoI’d be kind of surprised if they don’t watermark the content they generate. Just so they don’t train on their own slop.
- wenbin 1y agoMaybe some of them already embed some simple, secret marker to identify their own generated content. But people outside the organization wouldn’t know. And this still can’t prevent other companies from training models on synthetic data. Once synthetic data becomes pervasive, it’s inevitable that some of it will end up in the training process. Then it’ll be interesting to see how the information world evolves: AI-generated content built on synthetic data produced by other AIs. Over time, people may trust AI-generated content less and less.
- simianwords 1y agoWhy does this post not take into account the other side of it? AI has helped me become more grounded and correct in a lot of areas. Intricate questions are now easily answered in ChatGPT. Not only that - grok use in twitter works surprisingly well. Can some one really quantify the effect it has had in countering fake news? It is now way harder to spread fake news on X because a simple grok tag can counter it.
- crooked-v 1y agoHow do you know that those easy answers are reliably correct, though?
- simianwords 1y agoIt provides sources that you can look up if you are using it in the app. But in twitter: the simple fact that a lot of people don't seem to understand is that AI is actually really factually accurate - way more than you can expect from a normal person. I have seen very very few egregious errors from Grok and most of the factually incorrect posts seem to be caught by it. For the ones that I Grok was incorrect - I verified it myself and I was wrong and Grok was right. It turns out that Grok is actually really reliable. But the limitations are that it can't fact check extremely niche topics or intricate posts. But 90% of the cases it does catch them. Edit: it also makes cute errors like this sometimes https://www.timesnownews.com/world/us/us-buzz/grok-ai-mistakes-charlie-kirk-and-candace-owens-for-tom-holland-and-zendaya-article-152796224 https://www.timesnownews.com/world/us/us-buzz/grok-ai-mistak...
- lm28469 1y agoYeah it also makes cute errors praising Hitler &co. It seems like it's a fairly objective and factual bot as long as you're leaning on the same side as musk
- simianwords 1y agoCan you show more examples where it fails other than the obvious error that was admitted by them and fixed quickly? Of course this is just anecdotal - you will be able to find some bad examples but I'm still interested to know what you consider as errors.
- wwalker2112 1y agoWho or what is determining what is false or true in this study? I'm just as suspicious of this study as I am with the news.
- erefsg 1y ago[dead]
- 38054905485 1y ago[flagged]