What can artificial intelligence tell us about ourselves? Despite the growing urge to barricade myself in a log cabin on the side of a mountain, 300 miles away from the nearest circuit board, and answer “The day a robot tells me anything is the day I die,” it turns out A.I. can reveal a lot about the mysterious and murky history of human evolution. Although like all good mysteries, the more information we get, the murkier it becomes. According to an article published in Nature Communications, scientists from the Institute of Evolutionary Biology (IBE), the Centro Nacional de Análisis Genómico (CNAG-CRG) of the Centre for Genomic Regulation (CRG), and the University of Tartu have successfully used artificial intelligence deep-learning algorithms to identify the existence of a third, unknown, human species in the genome of Asian individuals.
We know that modern humans interbred with Neanderthals in Eurasia and Denisovens in Asia and Oceania, but scientists have theorized that there may have been a third species of hominid that made its way into the modern human genome. Now, thanks to these deep learning algorithms, that has been confirmed. The identity of third species is still a mystery, however, but scientists believe this species may have been a hybrid species of Neanderthal and Denisoven. This would bolster the argument that the remains of a Neanderthal-Denisovan hybrid found this past summer in a cave in Denisova was not an isolated incident, but evidence of an ongoing evolutionary progression. According to Mayukh Mondal, a researcher with the University of Tartu:
“Our theory coincides with the hybrid specimen discovered recently in Denisova, although as yet we cannot rule out other possibilities.”
The current model of human evolution holds that some 80,000 years ago, modern humans left the African Continent and began their migration throughout the rest of the globe. Up until 40,000 years ago modern humans co-existed with the other known species of hominids, Neanderthals and Denisovans. Recent studies have determined that these other species of hominids interbred with humans, giving us the weird hominid casserole that all of us beautiful people are today. While the recent findings don’t change that, it reminds us that the whole picture is quite a bit more complicated than we’ve previously been able to determine.
While the theory of a third hominid species has existed for a while, it was only through deep learning algorithms that scientists were able to prove it. According to Òscar Lao, principal investigator at the CNAG-CRG:
“[Deep learning] is an algorithm that imitates the way in which the nervous system of mammals works, with different artificial neurons that specialise and learn to detect, in data, patterns that are important for performing a given task. We have used this property to get the algorithm to learn to predict human demographics using genomes obtained through hundreds of thousands of simulations. Whenever we run a simulation we are travelling along a possible path in the history of humankind. Of all simulations, deep learning allows us to observe what makes the ancestral puzzle fit together.”
This is the first time that deep learning has been able to explain part of human history, and researchers are excited for the implications this has in biology, genetics, and evolution.
I’ll bet good money that somewhere down the line this, A.I. stuff is going to start feeding us disinformation. Sooner or later there’s going to be a journal article titled “Deep Learning Algorithms Determine Humans Were Born to Serve the Great Machine,” and you’ll be wishing you had the foresight to build a sweet cabin.