Fact Check: Did AI Predict Humans Evolving into Cyborgs in Viral Video?

A 39-second clip of an AI-generated animation has amassed over 11.6 million views on Twitter.

The rather disconcerting footage has been interpreted by some as a disturbing prophecy of the end result of human evolution, tracing the species from our chimpanzee ancestors to the present day and beyond, through to a cyborg dystopia.

But, as the artist behind this work told Newsweek, the real story behind the video is a little more complicated.

The Claim

On October 18, the clip was shared by Countere Magazine on Twitter with the caption: "What happens when you ask an AI to generate 'Human Evolution'. This is terrifying."

The video has been liked over 250,000 times and has received 53,800 retweets.

Others responded in shock and awe, tying the video to pre-existing narratives and theories of artificial intelligence intent on destroying humanity.

"Simulation theory confirmed. We are Lines," one Redditor responded.

"This is alarming," a Twitter user wrote.

The Facts

Based on the magazine's caption, readers would be forgiven for assuming that the AI had generated the lineage with a single prompt: human evolution. Under this assumption, it would seem the AI knew something sinister about the future of our species that we are yet to learn.

But Fabio Comparelli, the man behind the video, is keen to set the record straight.

"People are wrong when they say 'here is what an AI has imagined for the evolution of man'," Comparelli told Newsweek. "I clearly influenced the AI to generate the images I had in mind from the "prompt" text-based description at several key stages of the animation."

"It is me who decides from A to Z the theme and the description of each image, starting from the monkey, the prehistoric man, the medieval man, the modern man, the cyborg, the alien until the singularity where we return to the stage of raw information.

"The AI has no intelligence in itself apart from going to draw images from a huge database to offer me visuals in relation to the description of the image that I am asking for."

Comparelli created the video using Stable Diffusion, an open source model developed by Stability AI, which uses machine learning to produce images based on text prompts. The AI suggested thousands of computer-generated images based on each description. From these, the artist selected 900 images to make up the frames for the final animation.

"My goal is to do high-impact visual work that inspires others and sometimes delivers messages," he said.

The Italian-born artist only began working with AI three months ago and is already concerned about how people might view or interpret this relatively novel media.

"It is necessary to educate people on these new tools and not mix things up," he said. "It must be understood that the results of the AI are only a reflection of the desires of its creator.

"So many people have re-posted my work already with misleading titles...It's completely wrong and not respectful of my work."

The Ruling

False.

This animation does not depict an AI's "prediction" for the future of humanity. It was created by digital artist Fabio Comparelli with the help of AI-driven image-generating software. After receiving a series of text prompts, the algorithm had created hundreds of the individual frames, some of which were then handpicked by the artist.

This footage does not depict a dark prophecy but rather an innovative and striking digital artwork, created with the help of machine learning.

FACT CHECK BY NEWSWEEK