Jeremy Carrasco makes videos under the handle @showtoolsai on TikTok, advocating for AI video culture and highlighting what you can look for to see if the videos in your feed are AI-generated.
AILSA CHANG, HOST:
AI is everywhere, right? You probably see it all over your social media feeds. Remember those cute bunnies jumping on a trampoline? Fake. And now, because there are so many fake videos, there are now AI-based video reviews.
(SOUNDBITE OF ARCHIVED RECORDING)
JEREMY CARRASCO: In this video, different AI models. I think it’s Kling, which is a Chinese model, because there is Chinese text on the ceiling. And when he runs his hand through his hair, his little finger is completely up. They sell research-based chemicals, skin care products. There’s no way this is real.
CHANG: It’s Jeremy Carrasco. He’s a video producer who has spent years finding errors in real videos. He now takes a close look at AI. And on TikTok and Instagram, he teaches people how to spot fakes. Carrasco says detecting counterfeits has become much more difficult as AI models have improved. Missing limbs and fingers are no longer a sure sign. But the models are not yet perfect.
CARRASCO: For example, there are AI scams involving people sitting in hospitals and asking for money. But if you send these images to a doctor, he will realize that the equipment is completely wrong or half-made. If you watch a video of a climber, the ropes will be dangerous or not attached to anything.
CHANG: (Laughter).
CARRASCO: And as a casual observer, you might not know that. But the good news is that you have other superpowers or other expertise in one area that can be useful in other situations.
CHANG: So even if the scene looks realistic, your own logic might make you believe it’s fake. He says even something like camera angle can help you think critically about what you’re seeing.
CARRASCO: Absolutely. The main difference between these videos and our real videos is that there is no real person behind the camera. Most of the time the camera movement doesn’t make sense. And you just have to step back and ask yourself: who would have filmed this and why? There were lots of viral videos of dogs saving babies from falling shelves while you had a perfectly positioned camera.
CHANG: Yes.
CARRASCO: You know, that wouldn’t happen in real life. And if it looks like someone is holding a cell phone and a baby is plugging something into an outlet, that person would have put the film down to help, you know?
CHANG: (Laughs) That’s right, that’s true. Or like a photo of a cat jumping on someone’s bed in the middle of the night and catching a snake. For example, why would you have a security camera on yourself while you sleep at night?
CARRASCO: Exactly. I mean, not all of them are that obvious. But I’ll tell you, if you can catch the obvious ones – and pet videos are awesome because pet videos appear on all of our feeds. If you can spot these obvious things and learn from them from other details, then you’re well on your way to spotting them better overall.
CHANG: Well, I saw that TikTok has a policy that requires creators to label all AI-generated content that contains realistic images, audio and video. It’s politics. How reliable is this policy? And how prevalent are such policies among social media companies these days?
CARRASCO: Very unreliable. And they have some detection. But detection seems to be based more on descriptions or hashtags rather than actual analysis of the video. From what I can tell, no social media platform takes detection or labeling very seriously.
CHANG: Okay. Well, that brings me to my next question. Some of your videos talk about technical watermarks that AI generators embed into their content. But are these reliable indicators? I mean, how are these watermarks visible even to the casual consumer who encounters these videos?
CARRASCO: Yes, there are visible watermarks and invisible watermarks. So AI companies like Google are trying to introduce invisible watermarks embedded in pixels. But if you give Google a photo that’s AI-generated but not made by Google, they won’t be able to tell you if it’s AI.
CHANG: That’s right.
CARRASCO: So there is currently no good, reliable, centralized solution. But there are certainly a few promising ones that will hopefully be implemented soon.
CHANG: I mean, God, I’m talking to you right now and I’m thinking about all these methods you can use to spot fakes, you know, what I feel like is getting lost here is just the spontaneous pleasure that you would otherwise feel when you come across a funny video.
CARRASCO: Yeah.
CHANG: And I’m wondering, how do you think that hurts all these video creators who are creating real things?
CARRASCO: I agree with you. This is the saddest thing, because the best way to spot AI videos is to feel that sense of wonder that you may have felt…
CHANG: Yeah.
CARRASCO: …And you have to ask, wait, is that because it’s AI now?
CHANG: Yeah.
CARRASCO: And what I would say is there are a lot of real creators who still give you that feeling of pleasure. And you can just follow them and know that you can trust them not to screw up, because a lot of real creators really have no interest in creating AI videos. And AI accounts are only there to create AI. So I don’t think that sense of wonder has to go away. You just need to know where to look for it and not rely on the algorithms of social media platforms to simply present it to you.
CHANG: This is Jeremy Carrasco, AI literacy educator. Thank you very much for your ideas.
CARRASCO: Thank you, Ailsa.
Copyright © 2025 NPR. All rights reserved. Visit the terms of use and permissions pages on our website at www.npr.org for more information.
The accuracy and availability of NPR transcripts may vary. The text of the transcript may be edited to correct errors or match updates to the audio. Audio on npr.org may be edited after its original broadcast or publication. The authoritative record of NPR’s programming is the audio recording.
