“Most people think of robots as made of metals and ceramics but it’s not so much what a robot is made from but what it does, which is act on its own on behalf of people,” Bongard told CNN. “... The AI didn’t program these machines in the way we usually think about writing code. It shaped and sculpted and came up with this Pac-Man shape.”

Advertisement

Bongard told CNN that the “shape is, in essence, the program” and “influences how the xenobots behave to amplify this incredibly surprising process.”

Bongard told Gizmodo in an email that frog cells were used because it is one of the most common organisms used in biological studies. Levin and another biologist on the team, Douglas Blackiston, also have extensive experience working with frog tissue. Bongard explained that the team’s previous research into inducing the xenobots into specific behavior led to the discovery they could replicate themselves.

Advertisement

“In our first experiment from January 2020, we included frog heart muscle tissue into the xenobots and showed that could shuffle, slowly, across the bottom of a Petri dish,” Bongard told Gizmodo. “In a second paper from March 2021, we showed that xenobots can grow small hairs called cilia on their outer surface. They beat these cilia to swim, which results in faster movement than walking through water. We also showed that we could get the bots to ‘See’, ‘remember’, ‘come back’, and ‘tell’: the xenobots were induced to glow green.”

“When they come into contact with blue light, meant to represent something of interest to humans in their environment, they would permanently switch to glowing red,” Bongard added. “By counting the red bots at the end of the experience, we could tell how many bots had ‘seen’ the blue light. We also showed that a randomly-moving xenobot swarm would cause pellets in their environment to be pushed into piles. This was part of the inspiration for this current work... This led to the idea of replacing the pellets with individual cells, to see what would happen.”

Advertisement

Kinematic replication has been known to occur at the molecular level, but Bongard told Gizmodo it was never observed or believed to occur in organisms. According to the study, researchers verified that the xenobots, not “fluid dynamics and self-assembly,” were responsible for the replication after observing that the stem cells weren’t spontaneously combining in the xenobots’ absence.

In the study, the researchers wrote that kinematic replication and the spontaneous self-replication of the xenobots could help explain the origins of life on Earth. They wrote that more research could advance the amyloid world hypothesis, which “posits that self-assembling peptides were the first molecular entity capable of self-replication, and would thus represent the earliest stage in the evolution of life, predating even the RNA world.” The study could also contribute to the understanding of “how self-amplifying processes can emerge spontaneously, in new ways and in new forms, in abiotic, cellular, or biohybrid machines,” they added. On their website, the team speculates that xenobots could contribute to understanding of cell biology and eventually lead to advancements in regenerative medicine.

Advertisement

There’s really no telling what future xenobots might be used for, Bongard said. “It’s impossible to know what applications a very early-stage technology like xenobots will have,” Bongard wrote. “All we can do is consider the advantages this technology has over traditional robots, which is that they are small, biodegradable, and happy in water.”

“This means that, with the right regulation in place, they may operate in closed environments: they may be able to inspect plant roots in vertical farms, facilitate cultured meat production, or lower the cost of producing fresh water in desalination facilities,” Bongard said.

Advertisement

Levin told Gizmodo that possible applications for xenobots may arise in several areas. One is “useful, specific synthetic living machines (to do work in the body, in vitro for sculpting of tissues for transplants, in production facilities/plants, in the environment, in exploration, etc.),” he wrote, while another is “using the xenobots as a sandbox in which to learn how to convince groups of cells to build what we want them to build—once we can do that reliably, we will be able to have really transformative regenerative medicine for birth defects, cancer, traumatic injury, aging, etc. All those situations can be addressed once we understand how to stimulate cells to use their collective problem-solving ability to make the organs and tissues we want them to make.”

Levin added the xenobots might help scientists “better understand and control the goals and behavior of swarms of active agents—in this case cells, but those same lessons will help us make sure that the Internet of Things, swarm robotics, and many other technologies actually have beneficial outcomes.”