You might have missed the debut of the Xenobots last year when the world was falling apart, but they made quite a splash in the science and tech community. These Pac-Man-shaped synthetic organisms designed by supercomputers can organize into larger groups and be programmed to fulfill specific functions. They’re certainly not robots in the traditional sense, but they’re also too artificial to qualify as typical living organisms. They’re part cell, part machine, and completely one-of-a-kind.
As if all of that wasn’t already wild enough, the inventors of the Xenobots have just added one more trick up their creations’ sleeves—the ability to self-replicate. The implications of the new breakthrough, published Monday in the journal PNAS, are enormous. Scientists were already jazzed about the idea of one day using Xenobots for clinical applications and for cleaning up the environment. But the ability to replicate means this technology, if controlled, could be allowed to fulfill those tasks with more autonomy and on a larger scale.
“We found Xenobots that walk. We found Xenobots that swim. And now, in this study, we've found Xenobots that kinematically replicate,” Joshua Bongard, a computer scientist and co-author of the new study, said in a press release. “What else is out there?”
Xenobots are basically large structures of thousands of modified skin and heart cells from frogs. The skin cells provide rigid support while the heart cells act as small motors to let the Xenobot move forward. When the 1-millimeter-wide Xenobots were first unveiled to the world last year, scientists were exhilarated by their ability to swim out and self-assemble into larger tissues. Able to sense their environment, they could be programmed to push tiny objects around and transport payloads to different locations. They could survive for weeks without food or nutrients, and even heal themselves.
While scientists have mostly just used Xenobots so far to learn how cells organize into complex structures, many have suggested that they might be programmed to clear microplastics out of bodies of water, help deliver targeted drugs to specific locations in patients’ bodies, or build new tissues or organs to replace unhealthy ones.
But Xenobots won’t have much utility if researchers have to tediously make them one by one. Having the Xenobots replicate on their own would make it vastly easier to deploy them to do something useful in the real world.
Most complex cells in biology replicate by splitting in two, budding, or giving birth. But in chemistry, scientists have observed non-living molecules replicate by building copies of themselves out of ingredients found in the environment. That’s exactly what the new Xenobots do: They move around and collect frog skin cells that have not yet been aggregated into larger Xenobot bodies. Within five days, these piles of forgotten cells adhere to each other and become mobile, functional Xenobots themselves—capable of assembling new generations of Xenobots on their own as well.
There’s risk involved in creating synthetic living things that can simply generate new copies of themselves autonomously (see: every major sci-fi work about robots taking over the world). Before any kind of real-world applications can be spun out from Xenobots, the Xenobot team wants to fully understand the replication process so they can control it to do only what is sanctioned. They don’t plan to drag their feet on this, pointing to problems like climate change as all the more reason to move quickly.
“The speed at which we can produce solutions matters deeply,” said Bongard. “We need to create technological solutions that grow at the same rate as the challenges we face.”