Frog cells build miniature neurobots

Tiny frog-cell robots have been around for a few years. But they were missing the thing that makes animals feel less like clockwork and more like agents — a nervous system. That changed in March 2026, when teams at Tufts University and Harvard’s Wyss Institute described “neurobots,” little living constructs that don’t just move through water but also grow simple neural networks inside themselves. ### What is a neurobot? A neurobot is a small, fully biological machine built from cells of the African clawed frog, *Xenopus laevis*. Earlier versions, called xenobots or biobots, came from embryonic skin tissue that naturally folded into tiny spheres covered in cilia — hairlike structures that beat in sync and push the body through water. Neurobots start from that same setup, and sent out axons and dendrites through the interior. ### Why is that a bigger deal than it sounds? Because this is not just “frog cells plus a few neurons.” The interesting part is self-organization. Nobody wired these things up like a circuit board. The neurons developed inside an evolutionarily weird body plan — a body no frog has ever had — and still formed functional networks. Basically, the experiment asks a deep question: if you give cells the right ingredients, how much structure can they invent on their own? ### How did the team know the neurons were doing anything? They checked both structure and activity. Under the microscope, the neurons extended branches toward one another and toward surface cells. Then calcium imaging showed neural activity inside the bots, which is a standard way to see whether neurons are firing. The neurobots also behaved differently from xenobots. That is the part that pushes this from a neat tissue-engineering trick into something closer to embodied control. ### Are these actually robots? In the normal metal-and-motors sense, no. But in the broader sense — a designed system that senses, moves, and does work — yes, kind of. They are engineered living systems. The body is biological. The motion is self-powered. The control is emerging from tissues rather than silicon. That is why people keep reaching for awkward words like biobot, xenobot, and now neurobot. The language is clumsy because the object itself sits between categories. ### What changed from xenobots to neurobots? Earlier xenobots could already move, heal after injury, and survive for about 9 to 10 days on stored embryonic nutrients. But they did all that without an internal nervous system. Neurobots add a new layer — internal signaling tissue that changes morphology, movement, and gene activity. The paper also reports transcriptomic shifts tied to nervous-system development. Neurons are