Digital fruit fly brain walks while human neurons play Doom in biological computer experiments

Digital fruit fly brain model walks and cleans its feelers

San Francisco startup Eon Systems claims that it has created the first digital simulation of a fruit fly brain that can control a virtual body and produce recognizable behaviors. We'd call this "huge if true" but the research papers behind it look solid. This is science-fictional stuff - but not the SF you might think. The "first multi-behavior brain upload" is based on several substantial pieces of previous work, including some hefty scientific research papers in reputable journals. Otherwise, we would have dismissed it. On first look, the story sets several of our alarm bells ringing (brain uploading, AI, virtual reality, the Singularity). However, on the face of it, it seems to hang together. The researchers at Eon Systems have taken several pre-existing components: a fruit fly brain scan, a tool for modelling neurons, a model of some of the fly's muscles and body, and a very simple virtual environment, connected them together and ran it. The team claims that the result displays some of the behavior of the real insect. One part is a whole-brain connectome of a fruit fly, Drosophila melanogaster - one of the classic laboratory organisms, and a species this vulture spent hours peering through binocular microscopes at university in the 1980s (back then, he could separate male from female anesthetized flies in a dish at dozens to hundreds per hour). Effectively, this is a 3D brain scan of a fly's brain, including 125,000 neurons and 50 million synapses. The scan is from an adult female fly. This is important because one of the reasons Drosophila is a model organism is that some of the parts of their larvae are unusually large and so easier to work with. Relatively speaking, maggots don't do much, so a maggot brain scan would be easier - but no, this was an adult fly. The connectome comes from the splendidly named Flywire effort. That was published in 2024, and the same year, a team demonstrated that it seemed to respond to some signals, described in Nature in the paper "A Drosophila computational brain model reveals sensorimotor processing." Eon Systems ran the brain scan in a spiking neural network simulator called Brian2, and no, that is not a typo. Brian2 is FOSS and the source code is on GitHub. Eon also drew upon the NeuroMechFly v2 model, as described in the paper "Simulating embodied sensorimotor control in adult Drosophila." So, it took a synapse-level fly brain scan. It linked the scan's 125,000 neurons back up to one another using an AI tool to reconnect its 50 million synapses. Then it ran the resulting dataset inside an existing tool for modeling networks of neurons. It connected this digital model fly brain up to another model representing some of a fruit fly's body: its legs, antennae, mouth parts, and so on. The resulting model was fed some inputs provided by the MuJoCo advanced physics simulation. A video accompanying Dr Alex Wissner-Gross's Substack article about the work shows the model fly moving. In summary, the model fly brain, controlling a model fly body, managed to walk around, stop and clean its antennae, and when the model was presented with a signal resembling the smell of sugar, it stopped and extended its proboscis. In other words, it stuck out its tongue for a lick. We took the story to someone with considerable knowledge of running models of neurons in silico, Dr Steve Furber, co-creator of the ARM processor whose subsequent research included creating the SpiNNaker 2 hardware. He told us: Now, to be fair, at least one of the comments on the Substack post is severely critical: We specifically asked Dr Furber about that, and he said: Is Eon Systems' work an uploaded fly, waking up in the fruit-fly Matrix? No. But it does demonstrate interconnecting a modeled fly brain with a modeled fly body and getting fly-like behaviour out of it, and that is pretty impressive. Dr Wissner-Gross has been writing and presenting about AI and related matters for some 15 years, so he is clearly highly motivated to spin this for all it's worth, but even so, it looks like an impressive result to us. There are a number of things left unstated about the results so far. One of them is how long it took. This is going to be an absolutely massive model, which means running it will be quite slow. We're absolutely certain that the model and the video are not running in real time. It could well be that to get a few seconds of fly-like behavior took weeks or months of computer time. We are confident that there won't be brain scans of flying insects acting as onboard drone pilots in the next couple of years. We're sure that researchers around the world will be scurrying to replicate this, and indeed, to find any problems and highlight them. Among other things, though, it might result in easier access to funding. This is the sort of demo that you can wave in front of an uncomprehending administrator to good effect. The resonances from science fiction are inescapable, but they might not be as positive as you might immediately think. For instance, you might be excited that this could be a measurable, if tiny, step towards brain uploading. Be careful what you wish for, and read our favorite thought experiment about human brain uploads in recent years, Lena. It's a superb short story presented in the form of a fake Wikipedia article, and it is a chilling read. (The title is taken from one of the reference images used in digital image processing. The eponymous model herself now wishes the image was retired.) For a much longer version, there's Neal Stephenson's recent novel Fall; or, Dodge in Hell. This has the approximate shape of a happy ending if you squint, but the real subtext is right there in the title. What it directly and closely reminded us of, though, is no coincidence because the author is the person who shared the research with us. It was a short story called Lobsters by Charles Stross. This later became part of his 2005 novel Accelerando, which tells the story of humanity undergoing a technological singularity. A lot of folks in the tech world enjoyed this book hugely (it has 22,431 ratings and 1,448 reviews on Goodreads), without noticing the gigantic genocidal atrocity happening in the background - which should tell you important things about tech bros. For us, the takeaway was that, as the Russian invasion of Ukraine drags into its 12th year, the Gaza war is nearly 30 months in, and the new 2026 Iran war continues to send oil prices skyward along with the smoke, we are rolling into William Gibson's "Jackpot" scenario: Through the 21st century, it kills 80 percent of humanity, and most other life on Earth. As it happens, it is accompanied by a period of vast scientific and technological advance. ®

A petri dish of human brain cells is currently playing Doom. Should we be worried?

Scientists in the US have uploaded a fruit fly to a computer simulation, while an Australian lab has taught neurons on a glass chip to play a 90s video game. How long before we are all living in a sci-fi movie? It sounds like the opening of a sci-fi film, but US scientists recently uploaded a copy of the brain of a living fly into a simulation. In San Francisco, biotechnology company Eon Systems created a virtual insect that knew how to walk, fly, groom and feed in its virtual environment. Researchers in Australia, meanwhile, have taught a petri dish containing 200,000 human brain cells to play the iconic 90s shooter Doom. One experiment has pushed a brain into a computer; the other has plugged a computer into brain cells. Both stories have been hailed as scientific breakthroughs, but have also sparked inevitable fears about the prospects of lab-grown humans and digital clones. Should we be concerned? It was Australian startup Cortical Labs in Melbourne that taught a dish of lab-grown neurons to play Pong in 2022. Now it has built what it describes as "the world's first code-deployable biological computer", running on living human tissue rather than silicon chips, which is happily playing the 1993 shooter Doom. "In computer-science nerd land, there's this obsession with getting Doom to run on everything, from calculators to microwaves," Hon Weng Chong, CEO of Cortical Labs, tells me over Zoom from Melbourne. "As soon as we managed to get Pong to work, the first thing people said was: 'When are you going to do Doom?'" The average human brain contains about 86bn neurons - roughly 430,000 petri dishes' worth. But how do you harvest 200,000 brain cells without resorting to a hacksaw and an ice-cream scoop? "They're my brain cells, actually - at least most of them are," Chong says proudly. "There's no scraping or brain extraction. It's a very cool technique that was developed by Professor Shinya Yamanaka, who won the Nobel prize in 2012." All you need is 10ml of blood (in this case Chong's), from which around 100 white blood cells can be harvested. These can then be reprogrammed into induced pluripotent stem cells (iPSCs) - the body's biological building blocks - which can then be reproduced exponentially. "Essentially we reverse the biological clock back to an embryonic state, induce them into neurons, and put them on a glass chip roughly the size of a 50p piece," explains Chong . "Because they're on a chip - and electricity is the common language between neurons and the computer system - we can interface with them and get them to play Doom." Cortical Labs conducted its Pong experiment in-house, but this time it reached out to 24-year-old Singaporean Sean Cole, who has just completed an MSc in artificial intelligence at the University of Sussex, and whose dad happens to be mates with its CEO. Cole wrote the code remotely, which the team then tested on their local machines. "I was a bit surprised it worked the first time around," he tells me over Zoom. So how does a petri dish of brain cells play Doom when it doesn't have any eyes? Or fingers? "We take a snapshot of the game with information like the player's health and the position of enemies, pass it through a neural network, convert it into numbers, and send the data," explains Cole. "This is called encoding - essentially turning the game state into signals the neurons can understand. The neurons then fire an output - move left, move right, walk forward, shoot or not shoot - which the system decodes and converts back into actions in the game." "If you think about how humans operate, we have information going into our retina, which is converted into electrical signals, processed in the brain, and then an output happens," adds Chong. "It's really no different from that." If a computer full of brain cells is playing a video game and making decisions, does that mean it's sentient? Or is it just behaving like the average Doom player? "People have different perceptions of what sentience is," says Cole. "I definitely don't think it's conscious. At first it didn't know how to move, aim or even shoot. Then it would shoot the first two enemies and stop - almost as if it was preserving itself. So it's definitely learning. We've managed to control a brain to learn in a very controlled environment. The next step could be something like Neuralink, where you inject a chip into the brain to train someone to learn a language faster." Exactly how the cells are learning to play the game is unclear. "We can hypothesise that it might involve things like the free energy principle - the idea that living systems act to minimise free energy - or Hebbian learning, where connections between neurons strengthen when they fire together." Could we ever use technology like this to instantly learn kung fu, as in The Matrix? "If we find a way to safely connect this technology to humans, that is kind of what the implications might be," Cole says. "A big concern would be: what if you override someone's memories?" While Chong says he'd like to try getting the neurons to play Pokémon next, the real future application here lies not in getting trays of human neurons to graduate to playing Minecraft or Grand Theft Auto, but in medicine. "People are looking at it from biomedical research angles, for disease modelling," he says. "Things like epilepsy, where drugs could be tested on neurons grown outside the body - not just to discover new drugs, but to tailor them at a personal level." Meanwhile, in San Francisco, where Eon Systems has scanned a fruit fly brain and recreated it as a virtual insect, the big scientific news is that the team have essentially recreated the creature's behavioural wiring. The digital insect already knew how to behave like a fly, without any training or prompting. This challenges a central assumption of modern AI: that intelligence must be acquired. In the case of the fly, much of its behaviour appears to be built-in. "The brain was scanned using electron microscopy. Our head of engineering led a project to emulate that brain, and now we've placed the emulated brain back into a body, so it can wander around a virtual world," Michael Andregg, CEO of Eon Systems, tells me. A fruit fly's brain comprises around 140,000 neurons - about five petri dishes' worth. The virtual fly has 87 joints and can do pretty much anything an actual fly can. But does it realise it's living in a simulation? "The fly probably knows something's off, because we're not simulating the environment with high fidelity," says Andregg. "We can't give very specific taste and smell cues - just that something smells sweet or tastes bitter, but there are no complex aromas." Brain emulation, Andregg suggests, could eventually allow humans "to flourish in a world with superintelligence. Our goal is to make the emulation and computed brain and body feel indistinguishable from the natural biochemical body and brain," he continues. "If it feels different, we've done something wrong." But we're still a long way from the upload-yourself-into-the-internet futures imagined in Devs or The Lawnmower Man, mainly because, in this case, the fly's brain had to be removed from the body first. "Scanning the body was too hard," says Andregg, which will probably reduce the waiting list for human volunteers willing to try out the technology. Chong, meanwhile, believes biological computing could achieve things that traditional computing has struggled with. "There's a thing called Moravec's paradox, which is well known in robotics: what humans find very difficult, computers find easy, and what computers find difficult, humans find easy," he says. "Abstract reasoning, mathematics and language are relatively recent in evolutionary terms, which is partly why computers excel at them. But motor control and probabilistic decision-making are things we've inherited through millions of years of evolution. Robots may be very good at solving maths problems, but we're still trying to build robots that can walk properly." Biological systems like the fruit fly simulation, he says, might eventually power robots, drones and other machines that need to navigate the messy unpredictability of the real world. For now, humanity's first biological computer is busy doing what humans have always done with new technology: playing games. And somewhere in Silicon Valley, a fruit fly is living its second life inside a computer, totally unaware that it is living in the insect Matrix.