As AI grows in popularity and more companies develop realistic video generation models such as Google’s Nano Banana, viewing AI-generated content is becoming the norm. While some of these AI videos are cute (and obviously digitally rendered), like a baby kangaroo trying to board a plane, many AI content is being used to spread fake news and mislead people. In today’s volatile political world, AI-generated videos can play a major role in manipulating public opinion, making the ability to determine whether content is authentic becoming increasingly important in daily life.
While some companies are adding tools to make AI-generated content easier to find, there are still many ways people can publish this type of content. Stories about deer or black bears breaking bathroom windows may seem unbelievable, but more serious AI-generated videos may be embarrassing to share. Fortunately, there are ways to tell if a video is generated by AI, such as unnatural movement or expressions, or discrepancies between audio and video.
AI-generated videos are easy to spot if you pay attention to the details
On social media platforms, you should first look for tags or labels that indicate the content is AI-generated. For example, Meta allowed users to tag AI-generated content and also created an Instagram-like feed for AI video slops. YouTube requires creators to tag realistic content with “altered or synthetic media,” and most Veo 3 AI videos now display a visible watermark. There’s nothing wrong with posting AI content, but it’s more important than ever to be as transparent as possible about your use of AI.
Another thing to look for when determining if AI is being used in a video is to look for unnatural movements, according to Capitol Institute of Technology. If the person in the video isn’t walking regularly or eating very strangely, it’s likely an AI. The third element to watch out for is objects that appear and disappear out of nowhere. This may be one of the red flags that AI is more likely to spot. For example, if something in your video blends into something else and suddenly becomes a completely different image.
These can be good signs of AI-generated content
Capitol Institute of Technology also adds that audio discrepancies in videos can be a sign of AI-generated content. This fourth indicator is also easy to spot when someone is speaking, but the audio doesn’t match the lip movements on screen. Inaccurate background noise that is similar but not exactly the same is the fifth sign that the video was generated by AI. Imagine you are watching a video that is supposed to depict an urban environment. However, when you turn the sound up, the background noise is not the chaotic city noise you would expect to hear.
Last but not least, the best way to avoid confusion when it comes to AI-generated content is to fact-check the information yourself. If you find any of the above issues and are still unsure, use trusted websites to search for content. If your video is important enough, you should be able to get confirmation from a trusted source in just a few clicks from a news article or official website.
