It’s getting harder and harder to tell what’s real and what’s AI-generated these days. Maybe you’ve seen a viral image of a celebrity doing something outlandish or a news clip that seems a little off. Whether it’s a picture of a world leader holding a sign with an outrageous message or a video of a celebrity seemingly saying something shocking, this type of content is everywhere. While some of these creations are harmless or obviously fake, others are disturbingly convincing, and that’s where misinformation can take hold.
Even if you think you’re tech-savvy, AI can still trick you. We’re living in a time where AI images and videos are even being used by political campaigns, influencers, and media outlets, sometimes unintentionally spreading false information. If you’ve ever been fooled by one, don’t worry, you’re not alone. The key is knowing how to spot the signs of AI-generated content.
How to spot AI-generated images
One of the easiest giveaways of an AI-generated image is the presence of anatomical mistakes. AI still struggles with human hands, teeth, and facial symmetry, so watch out for extra fingers, melted-looking teeth, or ears that don’t quite match. Other red flags include warped background details—such as text that doesn’t form real words—or unrealistic lighting and reflections.
For example, we asked an AI image generator to create a photo of two people at a restaurant. At first glance, everything looked fine. But if you look closely, you can notice both individuals teeth looking sketchy; also, see the woman's thumb? It's blending with her pizza. There are other details... Can you spot them?
These small inconsistencies can often give AI images away.
Source: Leonardo AI
If you’re unsure whether an image is AI-generated, you can check using tools like AI or Not, which scan images and determine whether they’re likely AI-made.
How to spot AI-generated video
AI-generated videos can range from eerie deepfakes to laughably bad animation. The unsettling AI-generated “Anderson Cooper, 4k original” is a great example—watch it, and you’ll quickly notice the unnatural movements, stiff mouth, and garbled speech. But when an AI video is well-made, spotting the clues requires a closer look.
One of the biggest indicators of an AI-generated video is unnatural blinking or facial distortions. If a person’s face seems to change shape slightly, their eyes don’t track correctly, or the shadows on their face don’t match the lighting, there’s a good chance AI is involved. Lip syncing can also be an issue—deepfake videos often have slightly unnatural mouth movements, especially in side profiles.
If you suspect a video is AI-generated, tools like Deepware Scanner can help analyze and detect deepfake technology.
How to detect AI-generated audio
AI voice synthesis has advanced so much that it’s nearly indistinguishable from real speech. AI-generated audio clips can mimic real voices with startling accuracy, making them a popular tool for misinformation, scams, and even celebrity impersonations.
The biggest giveaway for AI audio is in the cadence and emotion. AI-generated voices often sound slightly too even, with a lack of natural breathing or variation in pitch. Sometimes, they over-pronounce words or have awkward pauses in places where a human wouldn’t normally stop.
If you’re unsure, AI detection tools like ElevenLabs’ AI Speech Classifier can help verify authenticity.
How to identify AI-generated text
AI-generated text is trickier to spot, especially as language models get better at mimicking human writing. However, a few telltale signs can help you recognize it.
First, look for an overly polished yet somewhat generic tone. AI writing often sounds fluid but lacks depth or originality. Repetitive phrases, abrupt topic shifts, or a lack of real-world knowledge can also be red flags.
For example, I asked an AI to write a travel guide about Rome. The text looked fine at first, but then I noticed odd factual errors—like claiming the Colosseum was built in the 1800s and that tourists could visit ‘Vatican Mountain.’ These inaccuracies, combined with a lack of personal anecdotes or nuance, indicated it wasn’t written by a human.
If you’re uncertain whether a piece of text was AI-generated, tools like GPT Detector and Grammarly’s AI detector can analyze and identify suspicious patterns.
Stay skeptical
AI-generated content is only going to become more sophisticated, which means we all need to stay sharp. Whether it’s an image, video, audio clip, or text, taking a second to analyze the details can prevent you from falling for misinformation. When in doubt, zoom in, listen closely, fact-check sources, and use AI detection tools to confirm authenticity. With a little skepticism, you can stay one step ahead of the AI-generated deception.
