Eon Systems announces digital fly brain emulation — but it's not what it seems

San Francisco startup Eon Systems became the star of last week's AI hype on social media. The company published a video of a so-called "digital fly" — a virtual simulation that co-founder Alexander Wissner-Gross called "the world's first embodiment of whole-brain emulation demonstrating several behavioral responses." The post spread across X, picked up by AI hype accounts and thousands of enthusiastic commenters.

Eon Systems positions itself as a company moving toward "digital human intelligence." As an intermediate goal, it names full digital emulation of a mouse brain — and promises to achieve it within two years. That timeline is, to put it mildly, ambitious: a mouse brain contains about 70 million neurons with tens of trillions of synaptic connections, and no team in the world has yet come close to reproducing it fully on a computer.

What did the video actually show? Based on available materials, it is a simulation of insect behavior built on a neural model — not an upload of a real fly brain into a computer, as many viewers may have assumed. The brain of the fruit fly Drosophila contains about 140 thousand neurons. Its wiring diagram — the connectome — was fully mapped by scientists only in 2023, after years of work. Even with that map in hand, building a functional simulation of all behavior is a completely different task.

The Verge put its skepticism right in the headline: "This is not a fly uploaded to a computer." The outlet notes that the buzz around the Eon Systems video rested largely on the audience's misunderstanding: people reposted the clip without grasping what they were watching. It is a typical AI hype mechanic: an impressive phrase, viral spread, and only then a substantive review.

Such cases have happened in neuroscience before. In 2014, the OpenWorm project recreated the neural network of the worm Caenorhabditis elegans (302 neurons) and "uploaded" it into a Lego robot. Media exploded with headlines about "uploading consciousness." Scientists then spent a long time explaining that it was a simplified model of reflexes, not consciousness and not even full behavior. History, it seems, is repeating itself.

What this means for the industry as a whole: the line between real scientific progress and marketing narrative in neurotech is getting thinner. Startups attract attention and investment with bold claims, and a public raised on AGI news readily believes in breakthroughs. Healthy skepticism here is not pessimism — it is simply a demand for precise wording.

A true brain emulation, even of a fly, would remain one of science's greatest achievements. But until that happens, it is worth reading the fine print.