591
Trains cancelled over fake bridge collapse image (www.bbc.com)
submitted 8 months ago by fantawurstwasser@feddit.org to c/technology@lemmy.world
78 comments fedilink hide all child comments
top 50 comments
sorted by: hot top controversial new old
[-] MagicShel@lemmy.zip 293 points 8 months ago* (last edited 8 months ago)

A BBC journalist ran the image through an AI chatbot which identified key spots that may have been manipulated.

What the actual fuck? You couldn't spare someone to just go look at the fucking thing rather than asking ChatGPT to spin you a tale? What are we even doing here, BBC?

A photo taken by a BBC North West Tonight reporter showed the bridge is undamaged

So they did. Why are we talking about ChatGPT then? You could just leave that part out. It's useless. Obviously a fake photo has been manipulated. Why bother asking?

[-] Deestan@lemmy.world 101 points 8 months ago

I tried the image of this real actual road collapse: https://www.tv2.no/nyheter/innenriks/60-mennesker-isolert-etter-veiras/12875776

I told ChatGPT it was fake and asked it to explain why. It assured me I was a special boy asking valid questions and helpfully made up some claims.

[-] Atropos@lemmy.world 59 points 8 months ago

God damn I hate this tool.

Thanks for posting this, great example

[-] plantfanatic@sh.itjust.works 12 points 8 months ago* (last edited 8 months ago)

Wait, you’re surprised it did what you asked of it?

There’s a massive difference between asking if something is fake, and telling it it is and asking why.

A person would make the same type of guesses and explanations if given the same task.

All this is showing is, you and ALOT of other people just don’t know enough about AI to be able to have a conversation about it.

It even says “suggests” in it, it’s making no claim that it’s real or fake. The lack of basic comprehension is the issue here.

[-] sem@piefed.blahaj.zone 31 points 8 months ago

A person would have the agency to ask, " why do you think it's fake?"

[-] plantfanatic@sh.itjust.works 7 points 8 months ago* (last edited 8 months ago)

Why would it have to? It and the person doing the task already knows to do any task put in front of it. It’s one of a hundred photos for all it and the person knows.

You are extending context and instructions that doesn’t exist. The situation would be, both are doing whatever task is presented to them. A human asking would fail and be removed. They failed order number one.

You could also setup a situation where the ai and human were both capable of asking. The ai won’t do what it’s not asked, that’s the comprehension lacking.

[-] sem@piefed.blahaj.zone 18 points 8 months ago

When people use a conversational tool, they expect it to act human, which it INTENTIONALLY DOES but without the sanity of a real human.

load more comments (7 replies)
[-] Weslee@lemmy.world 14 points 8 months ago

I think if a person were asked to do the same they would actually look at the image and make genuine remarks, look at the points it has highlighted, the boxes are placed around random points and the references to those boxes are unrelated (ie. yellow talks about branches when there are no branches near the yellow box, red talks about bent guardrail when the red box on the guardrail is of an undamaged section)

It has just made up points that "sound correct", anyone actually looking at this can tell there is no intelligence behind this

load more comments (1 replies)
[-] Deestan@lemmy.world 7 points 8 months ago

Wait, you’re surprised it did what you asked of it?

No. Stop making things up to complain about. Or at least leave me out of it.

[-] plantfanatic@sh.itjust.works 6 points 8 months ago* (last edited 8 months ago)

Then what are doing? Complaining it did exactly what you instructed it to do?

What else did you expect?

I get circle jerking against ai is hip and fun, but this isn’t even one of the valid errors it makes. This is just pure human error lmfao.

[-] WhyJiffie@sh.itjust.works 11 points 8 months ago* (last edited 8 months ago)

clearly, they asked it a question that average joe would do, and has shown that again its full of overly confident lies. it did not just reinforce the original belief of the user that it is fake, but it also hallucinated there a bunch of professional-like statements that are false if you take the time to check them. most people won't check them though, and straight up believe what it just spit out and think "oh this is so smart! outrageous that people call me dumb for asking it life advice!"

load more comments (1 replies)
[-] IcyToes@sh.itjust.works 70 points 8 months ago

They needed time for their journalists to get there. They're too busy on the beaches counting migrant boat crossings.

[-] BanMe@lemmy.world 56 points 8 months ago

I am guessing the reporter wanted to remind people tools exist for this, however the reporter isn't tech savvy enough to realize ChatGPT isn't one of them.

[-] 9bananas@feddit.org 27 points 8 months ago* (last edited 8 months ago)

afaik, there actually aren't any reliable tools for this.

the highest accuracy rate I've seen reported for "AI detectors" is somewhere around 60%; barely better than a random guess...

edit: i think that way for text/LLM, to be fair.

kinda doubt images are much better though...happy to hear otherwise, if there are better ones!

[-] rockerface@lemmy.cafe 17 points 8 months ago

The problem is any AI detector can be used to train AI to fool it, if it's publicly available

[-] 9bananas@feddit.org 12 points 8 months ago* (last edited 8 months ago)

exactly!

using a "detector" is how (not all, but a lot of) AIs (LLMs, GenAI) are trained:

have one AI that's a "student", and one that's a "teacher" and pit them against one another until the student fools the teacher nearly 100% of the time. this is what's usually called "training" an AI.

one can do very funny things with this tech!

for anyone that wants to see this process in action, here's a great example:

Benn Jorda: Breaking The Creepy AI in Police Cameras

load more comments (1 replies)
[-] Wren@lemmy.today 19 points 8 months ago

My best guess is SEO. Journalism that mentions ChatGPT gets more hits. It might be they did use a specialist or specialized software and the editor was like "Say it was ChatGPT, otherwise people get confused, and we get more views. No one's going to fact check whether or not someone used ChatGPT."

That's just my wild, somewhat informed speculation.

[-] Railcar8095@lemmy.world 15 points 8 months ago

Devils advocate, AI might be an agent that detects tapering with a NLP frontend.

Not all AI is LLMs.

[-] MagicShel@lemmy.zip 36 points 8 months ago* (last edited 8 months ago)

A "chatbot" is not a specialized AI.

(I feel like maybe I need to put this boilerplate in every comment about AI, but I'd hate that.) I'm not against AI or even chatbots. They have their uses. This is not using them appropriately.

[-] Railcar8095@lemmy.world 10 points 8 months ago* (last edited 8 months ago)

A chatbot can be the user facing side of a specialized agent.

That's actually how original change bots were. Siri didn't know how to get the weather, it was able to classify the question as a weather question, parse time and location and which APIs to call on those cases.

[-] MagicShel@lemmy.zip 23 points 8 months ago* (last edited 8 months ago)

Okay I get you're playing devil's advocate here, but set that aside for a moment. Is it more likely that BBC has a specialized chatbot that orchestrates expert APIs including for analyzing photos, or that the reporter asked ChatGPT? Even in the unlikely event I'm wrong, what is the message to the audience? That ChatGPT can investigate just as well as BBC. Which may well be the case, but it oughtn't be.

My second point still stands. If you sent someone to look at the thing and it's fine, I can tell you the photo is fake or manipulated without even looking at the damn thing.

[-] Railcar8095@lemmy.world 3 points 8 months ago

It's not like BBC is a single person with no skill other than a driving license and at least one functional eye.

Hell, they don't even need to go, just call the local services.

For me it's most likely that they have a specialized tool than an LLM detecting correctly tampering with the photo.

But if you say it's unlikely you're wrong, then I must be wrong I guess.

[-] MagicShel@lemmy.zip 8 points 8 months ago

what is the message to the audience? That ChatGPT can investigate just as well as BBC.

What about this part?

Either it's irresponsible to use ChatGPT to analyze the photo or it's irresponsible to present to the reader that chatbots can do the job. Particularly when they've done the investigation the proper way.

Deliberate or not, they are encouraging Facebook conspiracy debates by people who lead AI to tell them a photo is fake and think that's just as valid as BBC reporting.

load more comments (10 replies)
[-] brbposting@sh.itjust.works 3 points 8 months ago

If the article were written 10 years ago I would’ve just assumed they had used something like:

https://fotoforensics.com/

load more comments (1 replies)
[-] Tuuktuuk@piefed.ee 10 points 8 months ago* (last edited 8 months ago)

There's hoping that the reporter then looked at the image and noticed, "oh, true! That's an obvious spot there!"

load more comments (2 replies)
[-] mavu@discuss.tchncs.de 90 points 8 months ago

A BBC journalist ran the image through an AI chatbot which identified key spots that may have been manipulated.

WTF?

Doesn't the fucking BBC have at least 1 or 2 experts for spotting fakes? RAN THROUGH AN AI CHATBOT?? SERIOUSLY??

[-] Pyr_Pressure@lemmy.ca 51 points 8 months ago

They have vibe journalists now

load more comments (1 replies)
[-] Blackmist@feddit.uk 13 points 8 months ago

They do, they have like a daily article debunking shit.

[-] bilgamesch@feddit.org 6 points 8 months ago

People need to get that with the proliferation of AI the only way to build credibility is not by using it for trust but to go the exact opposite way: Grab your shoes and go places. Make notes. Take images.

As AI permeates the digital space - a process that is unlikely to be reversed - everything that's human will need to get - figuratively speaking - analogue again.

[-] ZILtoid1991@lemmy.world 4 points 8 months ago

No, they only have transphobia experts.

[-] myplacedk@lemmy.world 4 points 8 months ago

I haven't read it, but it could be to demonstrate how easy it was to identify it as a fake, without the ressources of BBC.

[-] wieson@feddit.org 3 points 8 months ago

Pr because it was between 0 and 2 in the night. Still, as an author I wouldn't have mentioned it.

load more comments (1 replies)
[-] alexsantee@infosec.pub 78 points 8 months ago

It's a shame to see the journalist trusting an AI chat-bot to verify the trustworthiness of the image instead of asking a specialist. I feel like they should even have an AI detecting specialist in-house since we're moving to having more generative AI material everywhere

[-] Tuuktuuk@piefed.ee 4 points 8 months ago

If the part of the image that reveals the image was made by an AI is obvious enough, why contact a specialist? Of course, reporters should absolutely be trained to spot such things with their bare eyes without something telling them specifically where to look. But still, once the reporter can already see what's ridiculously wrong in the image, it would be waste of the specialist's time to call them to come look at the image.

[-] MagicShel@lemmy.zip 7 points 8 months ago

What does that chatbot add?

[-] azertyfun@sh.itjust.works 4 points 8 months ago

My guess is the same thing as "critics say [x]". The journalist has an obvious opinion but isn't allowed by their head of redaction to put it in, so to maintain the illusion of NeutTraLITy™©® they find a strawman to hold that opinion for them.

I guess now they don't even need to find a tweet with 3 likes to present a convenient quote from "critics" or "the public" or "internet commenters" or "sources", they can just ask ChatGPT to generate it for them. Either way any redaction where that kind of shit flies is not doing serious journalism.

load more comments (1 replies)
load more comments (1 replies)
[-] ExLisper@lemmy.curiana.net 42 points 8 months ago

WTF? Why nothing like this ever happened during Photoshop times? Are people just dumber now?

[-] mic_check_one_two@lemmy.dbzer0.com 33 points 8 months ago

Because the venn diagram of “people who would maliciously do something like this” and “people with good enough photoshop skills to make it look realistic” were nearly two separate circles. AI has added a third “people with access to AI image generators” circle, and it has a LOT of overlap with the second group simply because it is so large.

[-] ExLisper@lemmy.curiana.net 9 points 8 months ago

Really? I remember tons of nicely photoshoped pictures on Snopes. There was a lot of trolling by people with skills going on.

[-] Vitaly@feddit.uk 11 points 8 months ago* (last edited 8 months ago)

The thing is you actually need some skill to do it in Photoshop, but now every dumb fuck who knows how to read can do shit like this.

load more comments (1 replies)
[-] DragonOracleIX@lemmy.ml 6 points 8 months ago

It took skill to do this before. Hardly anyone with that level of skill and time would do this. Now the dumb idiots have access to that skillset because of AI doing all the work for them.

[-] abbiistabbii 19 points 8 months ago

For anyone outside the UK, the bridge in the picture is carrying the West Coast Mainline (WCML).

The UK basically has two major routes between Edinburgh and Glasgow (where most people live in Scotland) and London, the East Coast Mainline and the West Coast Mainline. They also connect several major cities and regions.

The person who posted this basically claimed that a bridge on one of the UK's busiest intercity rail routes had started to collapse, which is not something you say lightly. It's like saying all of New York's airports had shut down because of three co-incidental sinkholes.

[-] rami@ani.social 18 points 8 months ago

I'm surprised to see no one else mention that it only took them an hour and a half to get an inspection done, signed of on and the lines reopened? That seems pretty impressive for something as important as a rail bridge.

[-] leftzero@lemmy.dbzer0.com 20 points 8 months ago

I mean, it's the time to get an inspector off of bed, on the road, to the site, and for them to go “yup, bridge's still there” and call back...

load more comments (2 replies)
[-] Blackmist@feddit.uk 17 points 8 months ago

Wait until this shit starts an actual war.

[-] sircac@lemmy.world 14 points 8 months ago

It feels like a privilege escalation exploit: at a certain point the authority chain jumped from a random picture provided who knows where/when to a link in the chain that should be reliable enough to blindly trust in this subject.

[-] entwine@programming.dev 7 points 8 months ago

A BBC journalist ran the image through an AI chatbot which identified key spots that may have been manipulated.

This is terrifying. Does the BBC not have anyone on the team that understands why this does not, and will never work?

load more comments
view more: next ›
this post was submitted on 07 Dec 2025
591 points (100.0% liked)

Technology

87325 readers
3035 users here now

This is a most excellent place for technology news and articles.

Our Rules

  1. Follow the lemmy.world rules.
  2. Only tech related news or articles.
  3. Be excellent to each other!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
  9. Check for duplicates before posting, duplicates may be removed
  10. Accounts 7 days and younger will have their posts automatically removed.

Approved Bots

founded 3 years ago
MODERATORS
L3s@lemmy.world
enu@lemmy.world
technopagan@lemmy.world
L4s@lemmy.world
L3s@hackingne.ws