The Fly in the Machine
In September a fruit fly learned to play Minecraft, or at least its wiring diagram did. On 3 September, Cell published the first complete wiring diagram of an adult male fly’s central nervous system: brain, optic lobes and nerve cord, 166,700 neurons and some 125 million synapses, every one traced and checked. By the next day a Georgia Tech graduate student had wired it into Minecraft. Within weeks the fly had been pointed at Doom, taught to trade bitcoin and sent to a virtual fruit fly heaven. Two months on, the clips are still being reshared across social media as if the news broke this week.
Behind the jokes sits a quieter story, and Israel is closer to it than most people realize.
In July 2009, at TEDGlobal in Oxford, Henry Markram told his audience that a computer model of the human brain could be built within ten years, and that if it worked a hologram would come back to give the talk. By his own account, almost all the biological groundwork for his Blue Brain Project was done at the Weizmann Institute, where he worked from 1995 to 2002. His Israeli collaborator of many years, Idan Segev of the Hebrew University, set the ambition alongside the Manhattan Project, Apollo and the Human Genome Project. In 2013 the European Commission backed the vision with the Human Brain Project and roughly a billion euros. The deadline came and went. When the project wound down in 2023, it left useful research infrastructure behind, but no simulated human brain, and no hologram.
The fly was built the other way round. Nobody promised a mind. They promised a map, and took nearly two decades to deliver it. The nervous system was cut into 66 slabs. In each one, a focused beam of ions milled away a few nanometers of tissue, an electron microscope photographed the fresh surface, and the beam milled again. The cycle runs until there is nothing left to photograph. Software built with Google Research traced the neurons through the images; human experts at Janelia then went through the result and corrected it.
One approach began with a theory of the whole and tried to fill in the parts. The other began with the parts and declined to say anything about the whole until every connection was accounted for. The second is the one people are now playing with.
That contrast should matter to Israel, because Israel is very good at the first kind of promise. A country that built its global brand on startups, and that is now putting national money into artificial intelligence and quantum computing, lives on announcements about what will be possible in ten years. Some of those announcements are vision. Some are fundraising. The fly is a reminder that the projects which change a field are often the dull ones, where the timetable is set by the material rather than the press release.
The most interesting finding is in the paper’s title: sexual dimorphism. Because a complete female fly brain had already been mapped, the teams could lay the two diagrams side by side. What differs is where it branches. According to the Cambridge group, a male fly that smells another male responds with aggression, while a female responds to the same scent with courtship. Same input, rerouted downstream, opposite behavior.
It is hard to read that without thinking of Israel. In 2023 the judicial overhaul reached every Israeli in the same words. Supporters read it as an overdue correction to an unelected court. Opponents filled the streets week after week, reading it as the removal of the last check on a parliamentary majority. The text was the same for everyone. What differed was where each reader’s wiring sent it. The fly proves nothing about people, and anyone who claims otherwise is selling something. But it offers a useful humility: the gap between two readings of the same event often lies not in the event, nor even in the first impression, but at a junction deeper in the circuit that neither side can see.
There is also the matter of who gets to play. The data is free. Anyone with a laptop can download the fly, and the first person to make it famous was a student, not a laboratory. That openness is science at its best, and it is also an exposure. Work funded by the UK Medical Research Council and the Wellcome Trust, alongside an American institute and an American company, now belongs to anyone who wants it. As brain maps move from flies to mice, and one day towards people, the question of whether such data should be published to the world or held as national capability will stop being academic. Israel, which treats so much of its technical talent as strategic, will be tempted to guard. It should publish. A small country gains more from being the place where the maps are made than from hoarding the ones it already has.
A brain about half a millimeter across can now be flown, parked and argued over by strangers. A wiring diagram is not a mind; it records structure, not the living chemistry of a brain or anything it might experience, and the people who built it are careful to say so. But it is the most honest step towards understanding a mind that anyone has yet taken, precisely because it promised so little.
Go back to the microscope. When the last layer was milled and the last image taken, the chamber was empty. The block of tissue that had held the fly was gone, turned into data one sliver at a time. The fly no longer exists anywhere. It exists everywhere.
