Viewing fake animal footage on social media can gain millions of views online and influence the way some of us perceive wild animals, new research has found.
AI can now generate emotionally rich clips such as a grateful baby polar bear, a sad puppy at an animal shelter, and an incredibly colorful tropical bird that unfortunately doesn’t exist. These heartwarming fakes have attracted millions of viewers who love the idea of animals acting like humans. Even people who are knowledgeable about animals may not understand something.
Paper published in conservation biology It focuses on a video by researchers at the University of Córdoba in Spain that depicts a fabricated bond between humans and animals. The authors warn that some of these can give people a false sense of security around wild animals.
This AI-generated image of a baby sloth has been viewed over 265,000 times and shared over 3,000 times on X.
Digital Disneyization
The Disneyification of the natural world has been going on for a long time. An old photograph from Banff National Park shows a tourist in the 1950s pushing a black bear behind the wheel of his car for a photo. Modern parents still sometimes try to put their toddlers on a deer’s back as if they were Bambi. But engaging AI videos take anthropomorphism to another level, the authors point out.
A second paper published this week in the journal Nature Ecology and Evolution argues that AI-enhanced images submitted to citizen science platforms like iNaturalist can contaminate the data researchers collect to determine the range and behavior of specific animals. They cite the example of an image that appeared to show a red-winged blackbird, submitted to iNaturalist as a possible first sighting of the species in Brazil.
In this case, the poster used an AI algorithm to “enhance” his image. This made the Epaulette oriole, commonly found in Brazil, look like a red-winged blackbird. He wasn’t intentionally trying to deceive. He just wanted to improve his image.
The team was able to reproduce his results using AI editing tools. They warn that these platforms risk becoming useless if too many posters enhance their images to the point of being unrecognizable.
