You’re relaxing in the living room after a big holiday meal, when your uncle starts flipping the vertical video. “Did you see a cat snatch that snake out of a guy’s bed?” he asks.
Is it true? Is it fake? You are feeling a headache.
“Recklessness is creeping up on us,” said Mike Caulfield, co-author of the book. Verified: How to think straight, be less deceived, and make better decisions about who to trust online“It floods the area and at some point your mental abilities take over,”
But Caulfield and other experts say you don’t need to succumb to despair, at least not yet. Dos and Don’ts There are some simple steps you can use to evaluate the authenticity of what you see online.
Don’t assume everything is fake
With so much carelessness in our feeds, it’s easy to think Everything What you see online is fake. But this bias is just as dangerous as believing that everything you see is real, warned Colina Koltai, a senior investigator at Bellingcat, an organization specializing in open-source investigations.
Bystander videos remain an extremely important source of evidence for misdeeds by individuals and law enforcement. When people stop believing those videos, researchers call it the “liar’s dividend” because it makes it easier for bad actors to avoid responsibility by claiming that real events are faked.
“I think that’s one of the bigger risks when it comes to this kind of material,” Koltai said. “It’s not that anyone will believe the fake video, it’s that people won’t believe the real video.”
Koltai and others say it’s especially important to carefully consider videos that may provoke a strong emotional response or run counter to your beliefs. Real videos often contain complex situations that can evoke reactions that challenge our understanding of the world. That being said, many fake videos are designed to do just that in order to increase engagement.
Be sure to pay attention to some simple features of the video
AI-generated videos are already excellent and improving rapidly, said Hany Farid, a professor at the University of California, Berkeley, who studies manipulated media. He said that even experts can be deceived. “I’ve been doing this every day for a long time, and it’s really hard. It’s In fact difficult.”
But there are some pretty straightforward features that can tell you whether or not the content you’re watching might be AI. The biggest tip is the length of the video.
Most companies limit the length of AI videos “because creating these videos is very computationally expensive,” Farid said. Many videos are only 8-10 seconds long. Although it’s possible to piece together a longer video with short cuts, “when you see those little bite-sized videos, it’s a good sign that you should take a breather.”
Length isn’t the only thing; AI-generated videos frame their subject perfectly, Farid said. The video features the main characters prominently, and the action starts and stops cleanly, even though the video is short. That’s why the quiz video of a New York City police officer yelling at ICE agents is clearly fake.
“It almost has a professional look to it,” he said. The position of the camera may also be awkward—is it too close to the subjects of the ICE raid? Or does it move as smoothly as if it were on a gimbal to follow a running animal? They could be clues generated by AI.
Be sure to check the references
The characteristics of the video are important, Caulfield said, but the location in which it is being shared sometimes matters even more.
Checking where the video was posted and even just looking at the comments can provide powerful clues. For example, the second video of an ICE raid in the quiz came from the Reddit community of Chicago’s Logan Square neighborhood.
Similarly, see who has posted the video. If their feed contains other types of content besides immigration raids, it lends credence to the idea that they watched the raids. Caulfield said, “It may be easy to fake a video but it’s harder to get into a time machine and make yourself ten years of history by talking about Chicago hot dogs.”
If you’re unsure whether the video you’re watching was originally posted by the account you’re viewing, try a simple reverse image search on Google or another platform. Often such searches may turn up the original post, other videos of the same incident, or news reports that either confirm or refute the video. The Logan Square ICE raid and Moose eating popcorn were both reported by the media at the time of the incident.
In contrast, identifying an AI video is often as easy as looking carefully at the account that posted it. It’s common for accounts to identify their content as AI-generated in their profile descriptions, Koltai said. Even if they don’t, checking the comments can often reveal that many people believe a video is AI-generated.
Don’t feel like you have to share it, especially if you have doubts.
In the end, all three researchers agree that in an age where algorithms reward speed over accuracy, sharing doesn’t really matter.
Much of the AI content being shared online is meant to entice participation, Koltai said. “Their creators often offer financial incentives for you to like, comment and share, as this often results in them making more money,” he said.
When in doubt, the best thing may be to wait, Caulfield said. “You don’t have to be the first one to share, you can be the one who waits,” he said. Often within a few hours, an incident will be confirmed by video or news reports.
Many people may not think that it matters whether you share an AI video of rabbits jumping on a trampoline, or a cat snatching a snake from its owner’s bed, but experts agree that it does matter. When people are duped by AI videos, it leads to loss of trust in important videos.
“People are like oh, is it really that big a deal?” He said. But Koltai said everyone should be concerned. “If we’re unable to tell what’s real online and what’s not? To me that’s really incredibly dangerous.”
Hany Farid agrees: “Every single one of those likes, clicks, shares, engagements, you’re part of the problem at this point,” he said.
<a href