How to spot AI-generated historical images?
AI tools are increasingly used to create fake historical images and videos, prompting experts to warn about their deceptive nature and provide tips on how to recognize them.
Artificial intelligence technologies are being widely used to generate fake historical images and videos, which often go viral on social media. According to experts, this trend is no coincidence – AI makes it possible to recombine images of the past and create synthetic scenes that could have existed but are ultimately misleading.
Roland Meyer, a media scholar at the University of Zurich, believes that AI-generated historical images primarily reproduce existing stereotypes and clichés rather than providing a true window into the past. "They are images of our expectations – not a window into the past, but a mirror of our present," he says.
One way to spot AI-generated historical videos is to ask whether such a video could exist at all. For example, a video depicting Pompeii before the eruption is clearly fake because video technology did not exist at that time. Also, look for visual clues such as robotic movements, anachronisms, and watermarks indicating the use of AI tools.
Second, scrutinize the source. Where does the image come from, who posted it and for what purpose? Historian and AI specialist Sebastian Kubon reminds that images always reflect a particular perspective and are not objective truth.
Third, compare with verified historical sources such as archives, museums and academic experts. For instance, there is no authentic footage of the Chernobyl disaster, so any video claiming otherwise is fake.
Historian Jürgen Zimmerer emphasizes that history is always an approximation. "We approach historical events when there is no original film footage, and if you pretend there is, you make things seem clearer than they actually are," he warns. Therefore, it is important not to fall for AI-generated historical slop but to dig deeper into the real past.
