Worried that a gadget leak or AI headline might be fake? Here is how to spot fake tech news: check the source, look for missing context, run a reverse image search, and look for fact checks before you share.
You can find related guides across the technology news section.
| Item | Value |
|---|---|
| Pope Francis puffer jacket image went viral | Early 2023 |
| Age of the Midjourney tool when that image spread | Nine months |
| Groups in the media literacy study | Three conditions (specific tips, general tips, no tips) |
| Study published in Cognitive Research: Principles and Implications | 2025 |
| Reliability of AI detection tools | Not always accurate; a second opinion only |
The figures come from AAP FactCheck and a Network Science Institute publication listing. Details can change, so check the source before you rely on them.
Why do fake tech stories spread so fast?
AAP FactCheck says generative AI gives everyone the ability to spin up text and hyperrealistic images, videos and audio to order within seconds. That speed is what makes a convincing fake cheap to produce and easy to pass around. If the technology is new to you, a primer on what is artificial intelligence explains the basics.
A well-known example is the image of Pope Francis in a puffer jacket, which went viral in early 2023 and fooled plenty of social media users. The tool behind it, Midjourney, had only been around for nine months but had already reached a convincing level of realism. The picture was posted to Reddit by an artist, and from there it spread around the world.
Not every fake is a joke. AAP FactCheck reports that scammers and spammers make money from fake pictures of sand sculptures, wood carvings and homemade cakes that gather thousands of clicks and shares. The same tricks can be pointed at a leaked gadget photo or a doctored screenshot of a product announcement.
Start with the source and your own reaction
The foundations of media literacy include asking who published the content and what their motive might be. Before you look at pixels, look at the publisher.
- Does the story come from a trustworthy news or government source, or from an unfamiliar account with no history of reliability?
- Does the claim seem too incredible to be true? If so, it needs further investigation.
- Is the post trying to make you feel something strong before you have checked anything?
On that last point, visual communication expert Dr T.J. Thomson of RMIT University told AAP FactCheck that evocative visuals aren’t always false, but ones seeking to mislead are often emotive or shared with rousing accompanying text.
Reputable people sharing something should raise your confidence, but even professionals get it wrong sometimes. For a wider method that goes beyond images, see the guide on how to read tech news critically.
Check the image or video itself
A shrinking subset of AI images still carry visual clues. Letters and numbers are often garbled and unreadable, and textures and shadows have proven hard for computers to copy. With video, AAP FactCheck suggests scrubbing through frame by frame to see whether any anomalies jump out.
Do not rely on those clues alone. As AI models improve, it will become increasingly harder to tell a fake from a real image, so how the content was produced and shared matters as much as what you can see.
Context is the other trap. A real clip can mislead if it is cropped, selectively chosen or altered, and AAP FactCheck gives the example of a video of Kamala Harris that was slowed down to make it appear she was slurring her words.
A reverse image search is the quickest way to test context. Tools like Google Images and TinEye can find other versions of an image online, which shows whether it was published before with a different description. For a video, take a screenshot of key parts and search with that to see where else it appeared.
What AI detectors and labels can and cannot tell you
Online AI detection tools let you upload an image and get a verdict on whether a human or a computer made it. AAP FactCheck calls them a useful second opinion, but they are not always accurate, so they cannot be treated as a definitive source of truth.
Labels have limits too. In its article dated January 28, 2024, AAP FactCheck noted that some platforms, including Facebook and TikTok, had started labelling AI-generated or AI-edited content. Psychologist Associate Professor Amy Dawel of the Australian National University told AAP FactCheck that people who misuse AI to deceive are unlikely to add a label themselves.
Dr Thomson adds that a label does not work for every kind of AI content. Knowing AI was involved also does not mean the image was not made to mislead.
What the research says about media literacy tips
A preregistered study by Sean Guo, Briony Swire-Thompson and Xiaoqing Hu, published in Cognitive Research: Principles and Implications in 2025, tested whether tips help. Participants were randomly assigned to one of three conditions: specific tips, general tips, or no tips as a control.
Specific tips increased headline discernment between true and false information more than general tips. Both kinds of tips reduced belief in AI-generated visual misinformation compared with the control, and specific tips reduced it more.
There was a catch. Both specific and general tips also reduced belief in real headlines compared with the control, with no difference between them. Healthy scepticism should not turn into doubting everything.
A quick checklist before you share
- Pause and ask whether the story seems too incredible to be true.
- Check who published it and why they might want you to believe it.
- Look for garbled text, odd textures or frame-by-frame glitches, without trusting those clues alone.
- Run a reverse image search, using a screenshot if it is a video.
- Type a short description of the image into a search engine along with the phrase “fact check” to see whether fact-checkers have already looked at it.
- Treat detector results and labels as hints, not verdicts.
Frequently asked questions
Can I spot an AI image just by looking at it?
Show answer
Sometimes, but less often over time. AAP FactCheck says only a shrinking subset of AI images show visual clues, so checking the source and context matters more.
Are AI detection tools reliable?
Show answer
They can offer a second opinion, but AAP FactCheck says they are not always accurate and cannot be the final word.
Does an AI label mean a post is misleading?
Show answer
No. Dr Thomson says that knowing AI was involved in making or editing an image does not mean it was or was not created to mislead. Labels also do not cover every kind of AI content.
How do I find out if a claim has already been debunked?
Show answer
Type a short description of the image or video into a search engine along with the phrase “fact check”. Professional fact-checkers regularly investigate images and videos shared online.
