Some uses of artificial intelligence amaze us and others scare us, but this one could forever change how we understand life in the universe. A team of researchers led by the Carnegie Institution for Science in Washington (USA) has developed a machine learning technology capable of distinguishing biological samples from non-biological samples with an accuracy of 90%. Scientists say this test can be reliable in detecting traces of present or past life on other planets. They call it "the holy grail of astrobiology." "This analytical method has the potential to revolutionize the search for extraterrestrial life and deepen our understanding of both the origin and chemistry of the earliest life on Earth," says Robert Hazen, one of the directors of the study that was published this Monday. the journal 'Proceedings of the National Academy of Sciences' (PNAS). "It opens the way to the use of smart sensors in robotic spacecraft, landers and rovers to search for traces of life before samples return to Earth," he adds. The new test could have a first application: revealing the history of the samples collected by the SAM instrument of the Curiosity rover on Mars. According to Hazen, "we will have to modify our method to match the SAM protocols, but we may already have data available to determine if there are molecules on Mars from an organic Martian biosphere." Standard Related News Yes Avi Loeb, the Harvard astrophysicist who believes in extraterrestrials: "We are not prepared for contact with aliens" José Manuel Nieves He is the Harvard scientist who since 2017 has been committed to investigating the possibility that Oumuamua, the first interstellar object that crossed the Solar System, was part of an ancient alien ship. And now he has done it again with some remains found in the sea that he thinks have the same extraterrestrial origin Molecular patterns "The search for extraterrestrial life remains one of the most tantalizing efforts of modern science," says Jim Cleaves, of the Laboratory of Earth and Planets at Carnegie. For this reason, he considers that the new research can have many implications, although “there are three important conclusions: first, biochemistry differs from abiotic organic chemistry; second, we can look at samples from Mars and ancient Earth to see if they were ever alive; and third, it is likely that this new method can distinguish alternative biospheres from those of Earth, with significant implications for future astrobiological missions. The researchers demonstrated that AI can differentiate biotic from abiotic samples by detecting subtle differences within molecular patterns, as revealed by pyrolysis gas chromatography (which separates and identifies the components of a sample), and spectrometry analysis. mass (which determines the molecular weights of those components). The AI was trained with a large amount of multidimensional data from molecular analyzes of 134 known abiotic or carbon-rich biotic samples. With approximately 90% accuracy, the AI successfully identified samples that originated from living things (shells, teeth, bones, insects, leaves, rice, human hair and modern cells preserved in fine-grained rocks), altered remains of ancient life by geological processing (coal, oil, amber, and carbon-rich fossils), samples with abiotic origins (pure laboratory chemicals, such as amino acids), and carbon-rich meteorites. Origin of life The authors add that until now the origins of many ancient carbon-containing samples have been difficult to determine because collections of organic molecules, whether biotic or abiotic, tend to degrade over time. Surprisingly, despite significant deterioration and alterations, the new analytical method detected signs of biology preserved in some cases for hundreds of millions of years. «These results mean that it is possible that we could find a form of life from another planet, from another biosphere, even if it is very different from the life we know on Earth. And, if we find remains of life in other places, we will be able to know if life on Earth and other planets has a common or different origin," says Hazen. "Put another way, the method should be able to detect extraterrestrial biochemistry, as well as life on Earth. This is a big problem because it is relatively easy to detect the molecular biomarkers of life on Earth, but we cannot assume that extraterrestrial life will use DNA, amino acids, etc.," he adds. 3.5-billion-year-old black rocks from Western Australia According to the authors, the technique could soon solve a number of scientific mysteries on Earth, including the origin of Western Australia's 3.5-billion-year-old black sediments, rocks that have sparked a great scientific debate. Some believe they are the oldest fossil microbes on Earth, while others claim they lack signs of life. Other ancient rock samples from northern Canada, South Africa and China spark similar debates. MORE INFORMATION news No An ancient manuscript from more than 800 years ago records a stellar explosion that will be repeated in 2024 news No Samples from Bennu, the asteroid that could collide with our planet, are already on Earth AI could also contribute to others fields such as biology, paleontology and archaeology, such as discovering whether an ancient fossil cell had a nucleus or was photosynthetic, discriminating different types of wood in an archaeological site or analyzing charred remains. “It's like we're dipping our toes into the water of a vast ocean of possibilities,” says Hazen.