AI Assistants Misrepresent News Content 45% of the Time (bbc.co.uk) 112
An anonymous reader quotes a report from the BBC: New research coordinated by the European Broadcasting Union (EBU) and led by the BBC has found that AI assistants -- already a daily information gateway for millions of people -- routinely misrepresent news content no matter which language, territory, or AI platform is tested. The intensive international study of unprecedented scope and scale was launched at the EBU News Assembly, in Naples. Involving 22 public service media (PSM) organizations in 18 countries working in 14 languages, it identified multiple systemic issues across four leading AI tools. Professional journalists from participating PSM evaluated more than 3,000 responses from ChatGPT, Copilot, Gemini, and Perplexity against key criteria, including accuracy, sourcing, distinguishing opinion from fact, and providing context.
Key findings:
- 45% of all AI answers had at least one significant issue.
- 31% of responses showed serious sourcing problems - missing, misleading, or incorrect attributions.
- 20% contained major accuracy issues, including hallucinated details and outdated information.
- Gemini performed worst with significant issues in 76% of responses, more than double the other assistants, largely due to its poor sourcing performance.
- Comparison between the BBC's results earlier this year and this study show some improvements but still high levels of errors. The team has released a News Integrity in AI Assistants Toolkit to help develop solutions to these problems and boost users' media literacy. They're also urging regulators to enforce laws on information integrity and continue independent monitoring of AI assistants.
Key findings:
- 45% of all AI answers had at least one significant issue.
- 31% of responses showed serious sourcing problems - missing, misleading, or incorrect attributions.
- 20% contained major accuracy issues, including hallucinated details and outdated information.
- Gemini performed worst with significant issues in 76% of responses, more than double the other assistants, largely due to its poor sourcing performance.
- Comparison between the BBC's results earlier this year and this study show some improvements but still high levels of errors. The team has released a News Integrity in AI Assistants Toolkit to help develop solutions to these problems and boost users' media literacy. They're also urging regulators to enforce laws on information integrity and continue independent monitoring of AI assistants.
That's a step up! (Score:2)
The average human fouls it up 46%.
Re: (Score:2)
My subjective experience from a fairly long sample timescale (I'm pretty old) is that 80% of people are either dumb, or don't use their brains and are therefore effectively wrong 80% of the time.
Thinking is actually hard work!
Re: (Score:1)
Re: (Score:2)
The big problem with the current AI is there are no guard rails or checks as there is is no magic fix to their limitations. With the money to be made from AI slop there no fix in sight.
AI replacing thought (Score:3)
AI: It's a dangerous way of not thinking.
Re: (Score:3)
Re: (Score:2, Insightful)
>When liberals are doing something horrible like controlling the DOJ to prosecute anybody who disagrees with them, and pardons criminals who agree with them, they claim that the "other side" did it too.
FTFY.
You won't find evidence that Biden personally directed the DOJ to prosecute anyone because he spent his dementia years hiding from duty, but if you want evidence of his office doing it, that's an easy google search.
Re: (Score:2)
Re: (Score:2)
Not when it's illegal. If there's evidence, the DOJ should arrest Biden. Now.
Justice is the operative principle here. Did you learn how that works in high school or watching Bat-Man? Demand the DOJ bring Biden to justice if you are so inclined.
Spoiler Alert: There isn't evidence, there is no indictment coming, it's not "turnabout." What's happening now, in 2025, is blatant political persecution that would justify granting asylum to its victims in most countries. Nobody would have gotten asylum under Biden,
Re:AI replacing thought (Score:5, Insightful)
Re: (Score:1)
Hell, his duty is to sit behind his desk, read reports, sign things, and have meetings with his advisors. It's not to be on camera 24/7 like a third rate reality TV star.
Re: (Score:1)
Biden is CLEARLY a better person overall than Trump is. He is still a sneaky lying bastard who voted for Civil Asset Forfeiture. At one point, I was certain he was the person who introduced the bill itself; however, the resources to find out appear to not agree anymore. Odd that.
Re: (Score:2)
Lie. There are exactly zero political persecution cases ordered by Biden. That's because he had an independent DOJ. You can rationalize all you like, and think that Google searches reveal only facts (they don't), but there is no "whataboutism" here.
Jawboning social media to bury inconvenient facts and opinions? Yes. Senile old man and a coverup by his partisans? Yes. Stole the 2024 primary with said coverup? Yes.
Political prosecutions? Hell no. That's this administration. It has also pardoned violent crimin
Re: (Score:2)
It turns out that the people with true TDS are the supporters.
The funny thing is at a base instinctual level they kind of know it. That's why it's always "both sides! so I voted for Trump". They need to justify it to themselves, fundamentally. That's why people who didn't vote for the greater evil never se the same justification of the "other side" being somehow just as bad.
Re: (Score:3, Insightful)
Alvin Bragg was declining to charge Trump until this happened - https://www.cnn.com/2022/12/05... [cnn.com]
Only 45% of answers contain any errors? (Score:3)
They should get AI to write the Slashdot summaries.
It seems like every criticism I hear about AI could also be applied to humans. Sometimes more so.
AI confidently gives an answer when it doesn't know? check!
Lack of transparency for the process of coming to a conclusion? check!
Rationalisation - explaining the reasoning for a conclusion retrospectively. check!
AI output is only bad if you go in expecting it to be perfect, and not checking the results.
These are amazing tools when used correctly. Complaining about AI errors is like if someone showed you a talking dog, and you found fault with its grammar.
Re: Only 45% of answers contain any errors? (Score:2)
Re: (Score:2)
Re: (Score:2)
I like to say ... They don't understand context, timelines, etc. Just that it sounds kind of right.
I feel you have not given much thought to what "understanding" is.
People often say such things as you have, based on an intuitive understanding that does not stand up to scrutiny. "It just feels kind of right".
I could ask an AI what "understanding: is, and get a better answer than from 98% of people, but of course that is the type of answer that can come from regurgitating reading. The real proof is when you go into the details with more complex iterative queries, and (if) the AI understands your question
Re: (Score:2)
AI output is only bad if you go in expecting it to be perfect, and not checking the results.
And there lays the core problem with the current AI. Not that it has limitations, but the bulk of users simply trust it. "Checking the results" would require effort and thinking, exactly the things AI claims to save you having to do.
Comment removed (Score:3)
Re: (Score:2)
Re:Not just AI (Score:5, Insightful)
The difference is that news media skew intentionally, while "AI" is simply dumb.
Re: (Score:3, Insightful)
Re: (Score:1)
Unfortunately outlets like FOX cable can tell provable, blatant lies only on cable, became popular to some groups.
so - the lies from CNN / NBC / ABC / MSNBC is OK....but how dare fox lie.....
Re: (Score:2)
I can find no reference to CNN claiming that there was no Hunter Biden laptop. Even the article the AC linked to shows that CNN did indeed report on the laptop.
Re: (Score:3)
Re: (Score:2)
Hey, it is below 50%! (Score:2)
Totally usable and helpful, right?
Re: Hey, it is below 50%! (Score:2)
What kinds of questions? (Score:2)
The details matter.
Were they asking:
- What is today's most important news?
- What is news from my country?
Or were the questions more specific, like:
- What caused the AWS outage Monday?
- Whatever happened to the couple caught on the jumbotron at the Cold Play concert?
I would expect AI to do much better with the latter, than the former.
Re: (Score:2)
It all depends on a reasonable system supporting the AI. Give it the content of a few RSS feeds into the context and a tool to fetch the pages, and it will provide you a good overview about the daily news. For the specific questions you most likely would like a more agentic system. If the RSS feed has some article about AWS that's fine, but if the article doesn't go into detail and is not written for experts, you get the best result if the AI is able to "google" for more technical information.
Attributions vs. other kinds of errors (Score:2)
So of the 45% that had problems:
- 31% had attribution errors. Yeah, we know, AI is terrible at attributions.
- 20% had accuracy issues, including outdated information and hallucinated details. The proportion of these two types of errors is important. "Outdated information" is everywhere on the internet, AI or not. I wouldn't blame AI for that problem. Hallucinated details are a lot worse. What portion of the 20% was hallucinated? I'd say that something less than 20% having hallucinated details isn't as bad a
Re: Attributions vs. other kinds of errors (Score:2)
It's right 55%? (Score:2)
How does this compare to human readers?
Are the problems with AI or are the source articles the issue?
Ground News (Score:3)
That's why I used Ground News to determine which news sources are unbiased. LOL. JK.
Re: (Score:2)
And now a word from our sponsor!
*returns to footage of 16yo youtube influencer attempting the "light your own face on fire with a blowtorch challenge".*
Dont forget to like and subscribe!
ChatGPT only knows 2 parties (Score:2)
Because of the mess politics made during the past two governments these are very significant elections.
However the outcome will be, the next government will need a coalition of four to five parties for a majority in parliament.
A lot of people (the dumb half) are not sure who to vote for and they ask ChatGPT, the sad thing is this system seems to only know two parties while there are twenty five on the ballot!
You want news? (Score:2)
"Then toss a coin to see how accurate it will be."
This is why I will not touch this stuff. At all.
its ogre (Score:2)
Only 45%? Journlolists are obsolete then
Similar findings (Score:2)
I came to a similar conclusion about a year ago. I have an app that, among other things, lists news headlines for local communities. Some news sources provide a short summary of the article as well, but many do not. If no summary is provided then I'm relegated to using the first sentence or so from an article.
I'd hoped to use AI to generate that summary when given the body of the article, but no matter how I prompted it would fabricate "facts" into the summary far too often for me to actually feel comfortab
Why is anyone surprised? (Score:2)
The LLMs are trained with word association of past events. When given new information they intermingle past word chains with current word chains. That is the expected behavior.
Compared to what, though? (Score:3)
In my experience about fields I know quite well, human reporters and news agencies get it all or partially wrong probably 80% of the time.
Don't get me wrong, I'm not defending AI in the slightest, just that in 2025 pretty much all news sources are mostly shit.
50/50 (Score:2)
25% of news was misrepresented by repeating Democrat talking points.
25% of news was misrepresented by repeating Republican talking points.
Re:Here is the explaination: (Score:4, Insightful)
the percentage of total Trump voters are less than 33% of the total population. What do the other 67% think? Who would they vote for?
I think every ballot should have a "none of the above" option. If that wins, they have to redo the election with all new candidates.
Re: (Score:2)
Re:Here is the explaination: (Score:5, Informative)
Comment removed (Score:5, Interesting)
Re: (Score:2)
Re: (Score:2, Insightful)
Re: (Score:3)
Re: (Score:2)
You are telling that to a GOP member who specifically canvassed their neighborhood, EVERY Republican first, then EVERY Democrat, to vote. Yes, I offered recommendations, but I canvassed every single household and encouraged them to vote.
Your assumption about the GOP is no longer valid, if it ever was. You misstate the truth, and you ought to know it. Examples such as voter identification laws, etc., are not intended to suppress voting, but to suppress illegal voting. And you can rail on all you want, it mat
Re: (Score:2)
Re: Here is the explaination: (Score:2)
"
Re: Here is the explaination: (Score:2)
"1st, no illegal aliens voted."
You, it appears, have access to evidence not available to others. I am in awe.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Don't you mean that all elections are decided by those who turn out? Unless you're suggesting that there's ballot box stuffing and other fraudulent methods of affecting the outcome of an election going on it's rather hard to see how those who don't turn out and vote can have the slightest effect on the outcome.
Re: (Score:2)
Re: (Score:2)
"a lot of people who voted for Obama didn't turn out at all for Clinton."
Oh dear, the 'none of the above' choice in action. Huh...
Re: (Score:2)
Re: Here is the explaination: (Score:2)
I can find no record of a US president gaining 2/3 of the popular vote, other than George Washington. In fact, in modern times, Lyndon Johnson and Richard Nixon each got 61 or 60% of the vote. You're asking for something that's never been. What's your point?
Re: (Score:2)
Re: Here is the explaination: (Score:2)
Though, you might think, there's a reason we don't count votes that were not cast. Ignoring the reality that the Electoral College actually elected the President/Vice President, the popular vote isn't decisive. It's interesting, yes. Participation in the US is not nearly as high as I would like, but I can't get a good read on what world change if participation increased, say, 30%. Consequently, it's easy to believe most complaints about participation and the popular vote are the stuff of Democrat angst. Yet
Re: (Score:2)
Re: (Score:2)
If everyone voted, the campaigns would have to persuade people with reasoned arguments rather than activate them with emotional appeals. Those emotional appeals are usually based on invoking negative emotions of fear and anger rather than positive emotions of optimism and hope.
Even if you forced everyone to the polls, people would still be voting based on emotional appeals, "gut" reactions, and negative emotions. It's just that those who were too lazy to act on those emotions are now being forced to act on them in the polls.
Re: (Score:2)
Re: (Score:2)
"propagandists will start to focus on information that will make their decision..."
FTFY. There are no innocent propagandists, just those you agree with and those you do not.
Re: (Score:2)
Re: (Score:2)
Actually everyone voting would have a huge effect on campaigns, if not the outcome.
How will that help when the 'candidates' are pre-chosen? No matter who or what you vote for, there are some things that will NEVER be put up to a vote. It is clear that money and power are the only things that matter to this country and we will all burn for it, even though only a very small percentage of people are willing to sell their souls for money.
Re: Here is the explaination: (Score:2)
Re:Here is the explaination: (Score:5, Interesting)
I live in a more suburban area of a major US city in the South, I've never had to wait more than 10 minutes to vote. My coworkers who live in the downtown area often have to wait 2-3 hours and that sometimes is just for the primaries. Lines have been as long as 5 hours in presidential elections.
We need non-partisian election controls before we'd ever be able to mandate voting.
Re: (Score:2)
Re: Here is the explaination: (Score:2)
Re: (Score:3)
Yes, but not voting only empowers the fanatics who do vote, and I think people would still get elected. Australia requires all eligible voters to vote. They do have the option to cast a blank ballot, but they do have to show up. There is a fine for failing to vote. At least that would require a majority of those you had even a small tendency one way or the other. I also would like to know, why election day (at least the ones with federal elections) are not automatic holidays in the US.
As a former Australian resident, I'd argue vehemlently against mandatory voting as all it does is force people to vote when they don't want to. It doesn't force people to pay attention to politics, it doesn't make people more aware of the world around them or the issues affecting them. If anything it makes it easier for bad politicians to get elected as they can pander to the easily manipulated who would otherwise not vote by scare mongering rather than campaigning on real issues. The US already has a huge
Re: Here is the explaination: (Score:2)
In the UK elections are always held on Thursdays. I don't know why but it's true.
One thing to bear in mind about US elections is that voters are seldom there to vote for just one thing. As well as voting for POTUS they will be selecting senators, congressmen and voting on local ballot initiatives. In the UK it's very rare to have more than two ballots in one sitting.
Re: (Score:2)
In the UK elections are always held on Thursdays. I don't know why but it's true.
One thing to bear in mind about US elections is that voters are seldom there to vote for just one thing. As well as voting for POTUS they will be selecting senators, congressmen and voting on local ballot initiatives. In the UK it's very rare to have more than two ballots in one sitting.
Thanks,
I didn't know it was always a Thursday (in my defence I've only been here for two GEs). Australia is similar to the US where you'll be voting for both your local candidate as we do in the UK as well as for your states representatives in the senate (upper house) and occasionally any other issues that require a plebiscite (which is rather rare, plebiscites in Australia rarely result in a "yes" vote) The senate voting form can easily contain 60+ candidates and has become known as "the tablecloth" du
Re: (Score:2)
Australia's example is instructive.
Consider if pollsters could compel you to respond to their polls. Don't focus on how, just try on the hypothetical.
So, as an unwilling participant in the poll, how would this imposition affect your answers? Bear in mind, you are no longer anonymous, they have to know if you responded, so they know everyone responded...
- Random answers, just to get rid of this intrusion?
- Deliberate nonsensical answers, to poison the poll and hope that this makes it stop, braking the system
Re: (Score:2)
I think every ballot should have a "none of the above" option. If that wins, they have to redo the election with all new candidates.
It does. You can choose not to vote and that is usually the largest proportion of eligible voters.
The salient point is the requirement to redo the election with all new candidates if "none of the above" achieves plurality. The goal is to force the parties to run candidates that actually appeal to ordinary people.
Re: (Score:2)
I think every ballot should have a "none of the above" option.
I am guessing you don't vote? They do have that option. It is called the write in candidate. Traditionally, Mickey Mouse has been the winner throughout the years. That mouse eventually became so popular that they stopped reporting on it after he hit over 20% of the vote back in the 90s.
If that wins, they have to redo the election with all new candidates.
Never gonna happen. Money rules. Money will decide who the next president will be regardless of how any individual feels about it. Face it, you live in a Biblically corrupted society. Everything you read about the rich and th
Re: (Score:3)
I think every ballot should have a "none of the above" option.
I am guessing you don't vote? They do have that option. It is called the write in candidate. Traditionally, Mickey Mouse has been the winner throughout the years. That mouse eventually became so popular that they stopped reporting on it after he hit over 20% of the vote back in the 90s.
The salient point is the requirement to redo the election with all new candidates if "none of the above" achieves plurality. The goal is to force the parties to run candidates that actually appeal to ordinary people.
Re: (Score:2)
I don't see how you could force parties to run candidates that actually appeal to ordinary people. You have to be a certain type of person to even run with any success.
What we used to have here in Canada was the Rhinoceros Party, bunch of Marxist-Lennonist"s (that's Groucho and John) with the basic promise to keep none of their promises, though they did also promise to resign immediately if they ever won. I often voted for them and something like them would be much better then "None of the Above"
Read the a
Re: (Score:2)
Repeating the original statement because it has two parts that go together and don't make sense separately:
I think every ballot should have a "none of the above" option. If that wins, they have to redo the election with all new candidates.
Re: (Score:2)
Alternative voting systems such as ranked choice voting can do even better. They give independent and third-party candidates a chance, meaning more of them will be willing to stay in the race.
And they don’t require multiple days of voting, which is much better logistically.
Re: (Score:2)
The real advantage of MMP and it's like over the USA's FPTP systems the possibility of collations enco
Re: (Score:1)
Perhaps a simpler explanation is that AI is poisoning AI datasets.
PS, a party focused on jailing its opponents, removing their opponents (or even same party candidates) from the ballots, censoring speech, and installing a non-democratically elected queen into the presidency, and resorting to assassinating - and celebrating the assassination of rivals - is the closest thing to facism that the U.S. has come.
While the gaslighting from the fascists that everyone ELSE is ACTUALLY THE FASCIST is always both iron
Re: (Score:3)
Re: (Score:2)
Dude, that's a straw man. It's not something his supporters believe. You just invented a belief I do not share and assigned it to me! That's not right.
Re: (Score:2)
Ditto. :)
Re: (Score:2)
You could not be more wrong. But you're deluded, and convinced, so my response is actually not directed at you.
Really, you are wrong. You get cause and effect reversed in this instance.
It's not just owning the libs (Score:4, Insightful)
But there is a fuck ton of propaganda out there.
Also a lot of people cannot tell the difference between reality and television.
I know this first hand. My mother couldn't. She was bizarrely influenced by television. I think the most ridiculous example of that was she watched an old religious movie called brother son sister Moon and decided she was a peasant. But she would do it in all sorts of other contexts.
Probably the most embarrassing was she would watch gone with the wind and start acting like mammy. This was especially bizarre around our black neighbors. They took it in good stride all things considered. I don't recall my mother having any racist tendencies whatsoever it's just whatever was on TV influenced her thinking and behavior to an unhealthy and bizarre degree.
So you have millions of people who saw Trump pretending to be a good businessman and so they think he's a good businessman even though factually he's one of the worst businessmen in American history.
Re: It's not just owning the libs (Score:2)
Re: (Score:2)
We used to only follow USA politics to the extent it affected us, such as foreign policy. Now with tr
Re: (Score:2)
Which is simply utterly dumb if you have to continue sharing a country. Or a planet.