Viewpoints Explained: The Tiny Clues That Expose Artificial Intelligence Footage

This is Viewpoints Explained…

Go back two or three years ago and videos that were created by artificial intelligence were easy to spot. Today, they’re far more refined, making it harder to figure out what’s real and what’s not. For example, new video creation platforms like chat GTP’s Sora are able to quickly produce hyper-realistic humans that move and talk just like we do.

If you’re watching a video online, here are some easy tell’s that it’s probably not real. Experts say that these videos often show strange blinking patterns or eye movements that don’t track with what the rest of the face is doing. Hands are another giveaway. In many clips, people have the wrong number of fingers or the fingers themselves are too long or misshapen.

Weird lighting can also be a red flag. Some videos show shadows that don’t match the environment and shift in ways that don’t line up with the person’s movement. A-i can create weird text in the video by displaying warped letters, odd spellings or glitchy captions.
Beyond visual clues, it also helps to question the circumstances around a clip. Videos that appear out of nowhere, arrive just as a story is breaking or come from accounts with little activity deserve a second look. Check whether reputable, well-known newsrooms or local officials are reporting the same event – or if the clip seems to stand alone.

The post Viewpoints Explained: The Tiny Clues That Expose Artificial Intelligence Footage appeared first on Viewpoints Radio.