After designing these cells, the scientists were able to activate new sensors and motors on them and even produce them faster. But like all living cells, these robots hit a roadblock in death. A few days after they were manufactured, the cells inside the 0.7-mm organisms would begin to die ultimately leading to their death.

After designing these cells, the scientists were able to activate new sensors and motors on them and even produce them faster. But like all living cells, these robots hit a roadblock in death. A few days after they were manufactured, the cells inside the 0.7-mm organisms would begin to die ultimately leading to their death.

So, the team wondered if there was a way they could get the organism itself to reproduce. They posed the question to the Deep Green Supercomputer cluster at the University of Vermont in the form of an evolutionary algorithm. Billions of shapes were tested to determine what would allow these bots to replicate and after months of analysis, the supercomputer came up with a familiar idea: Pac-Man.

So scientists designed ‘parent’ xenobots in the shape of Pac-Man. Inside a petri-dish, these bots could then swim out to individual cells and gather hundreds of them at a time. After staying in parent xenobots’ Pac-Man-like mouths for a few days, these cells transformed into new “baby xenobots” that moved and acted as the parent.

“Then those parents built children, who built grandchildren, who built great-grandchildren, who built great-great-grandchildren,” said Sam Kriegman, the lead author of the study, explaining how the design gave rise to generations of xenobots. “These are frog cells replicating in a way that is very different from how frogs do it. No animal or plant known to science replicates in this way,” he added.

“People have thought for quite a long time that we’ve worked out all the ways that life can reproduce or replicate. But this is something that’s never been observed before,” said Douglas Blackiston, a scientist at Tufts University who assembled the ‘parent’ xenobots and is also a co-author of the study, published today in Proceedings of the National Academy of Sciences.

Professor Josh Bongard who leads the xenobot research at UVM said that understanding these systems will open doors to many new technological developments. Citing the timeline for the development of the COVID-19 vaccine, Bongard called it “exceedingly long” and said that using his research in the future, we could direct AI to make a biological tool that does specific functions for us.