QUICK question: did you have “self-replicating nanobots that could potentially consume the entire Earth” on your 2021 bingo card? I sure didn’t. Yet, here we are. Or so it seems.

As if there were not enough frightening indignities in the world to contend with — Covid-19, Republicans, who damned Mariah Carey Christmas songs — scientists have presented us with an entirely new horror. In a study published in the journal, “PNAS,” this past Monday, a team of researchers from the University of Vermont, Tufts University and Harvard University announced that “living robots” they created last year were now able to reproduce “in a way not seen in plants and animals.”

The artificial organisms, which are slightly smaller than a millimeter across, were created from the stem cells of the African clawed frog (Xenopus laevis), hence their name, “xenobots,” because the fact that they can move, work together in groups and heal themselves isn’t terrifying enough.

In the latest study, the researchers explained that the original form of the xenobots — spheres made up of about 3,000 cells each — could replicate, but only rarely, and so the objective was to improve their performance in this regard. To do this, the scientists handed the problem to an artificial intelligence (AI) program, which tested billions of different body shapes to make the xenobots more effective at replication. The AI program eventually settled on a C-shape, which is able to collect hundreds of stem cells into a bundle in its “mouth,” which emerges a few days later as an entirely new xenobot.

The lead author of the study is Josh Bongard, a computer science professor and robotics expert at the University of Vermont, and what he had to say about the project and its results only adds to the horror.

“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 said, displaying an altogether too trusting assumption of the critters’ motives.

“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,” he explained. “The shape is, in essence, the program.

“The shape influences how the xenobots behave to amplify this incredibly surprising process.”

The researchers said the technology is still in its early stages and does not have any practical application yet, but that the “combination of molecular biology and artificial intelligence could potentially be used in a host of tasks in the body and the environment,” such as collecting microplastics in the oceans, cleaning up oil spills and regenerative medicine.

The scientists also acknowledged that the development of self-replicating biotechnology “could spark concern.” They, however, offered reassurances that the project was completely safe as it is being conducted in a contained laboratory setting and is “regulated by ethics experts,” which is an extremely ironic thing to say about any project that is funded in part by the Defense Advanced Research

Projects Agency (Darpa), the spooky US agency that oversees the development of technology for military use.

Let’s run this around the block again, shall we? First, scientists extract some stem cells from a frog, put them in a suitable environment and watch as they become a completely new form of artificial organism able to act autonomously or cooperatively as circumstances dictate, which is close enough to a definition of rudimentary consciousness. Then, the scientists allow an AI program (which by definition can learn and adjust its behavior on its own) to find the most efficient form for the xenobots to reproduce on their own. The scientists follow these instructions, build some xenobots in that shape, and presto, the little bugs energetically set about building new versions of themselves in an “incredibly surprising process,” that the researchers did not expect, have never seen before, and do not completely understand.

In a project that is partly being backed by a black-bag US agency that exists solely to find new technologies that can be used as weapons.

Oh sure, this is totally fine. No way this can go wrong.

The ‘Gray Goo’

As the United States does not have a monopoly on smart people or lab equipment, the most likely thing to happen is that someone else will figure out how to produce xenobots for purposes that are not exactly “on behalf of people.” There is, however, another even more horrifying prospect, and while it still may be a remote possibility, the recent discoveries have made it a lot less unlikely.

It’s called “Gray Goo,” and it goes something like this. First, scientists create some form of self-replicating nanotechnology, such as the xenobots. As the xenobots are capable of a certain level of evolution, they adapt to whatever environment they are placed in to continue to harvest matter for the creation of new copies of themselves. For example, they may be used to consume an oil spill, converting the organic compounds of the oil to new xenobots. Once all the oil is consumed, however, the xenobots would keep going, and start consuming any available resources.

Let’s say our hypothetical xenobots are capable of producing one new “offspring” per hour, and we start off with 1 kilogram of xenobots. After 12 hours, that 1 kilogram has become 4 tons of xenobots; after a day, 16,777 tons; in 36 hours, about 68.8 million tons; and in two days, 281.5 billion tons. Within another few hours, the mass of xenobots will weigh more than the entire Earth, turning the surface of the planet into a miles-deep swamp of “gray goo.”

That scenario is, of course, the stuff of science fiction and dumb History Channel shows, but on the other hand, it is not a theoretical impossibility. Runaway xenobots or similar artificial organisms would almost certainly run out of things to “eat” well before they turned the planet into “gray goo,” but they still could cause serious problems; realistically, they would probably behave more like a crop infestation. Or a pandemic.

Maybe it’s just me, but I think that science can probably find better things to do than to flirt with that possibility.