Can AI Detect Signs of Life on Other Planets? To some extent, they already do. Sensors aboard spacecraft exploring other worlds have the ability to detect molecules indicative of alien life. However, it is known that organic molecules derived from biological processes degrade over time, making their presence difficult to detect with current technology.
A recently developed method based on artificial intelligence (AI) is capable of detecting biological signals, even in samples that are hundreds of millions of years old. Furthermore, according to new research, this method provides results with an accuracy of 90% (ref.).
Biomolecules vs. Abiotic Molecules
The new method is based on the premise that the chemical processes governing the formation and functioning of biomolecules fundamentally differ from those of abiotic molecules. Biomolecules, such as amino acids, retain information about the chemical processes that produced them. According to the new study, this likely applies to alien life as well.
On any world, life can produce and use greater quantities of select compounds for everyday functioning. This would differentiate them from abiotic systems, and these differences can be identified and quantified using artificial intelligence.
The team first trained the machine learning algorithm with 134 samples, including 59 biotic and 75 abiotic samples. Subsequently, for validation, the data were randomly divided into a training set and a test set. AI successfully identified signs of life in biotic samples from recent living organisms, as well as in ancient life forms preserved in some fossil fragments made of materials like coal, oil, and amber.
Tests on Earth
According to the study, the tool also identified abiotic samples that include chemicals like laboratory-created amino acids and carbon-rich meteorites. Almost immediately, the new artificial intelligence method could be used to study rocks dating back 3.5 billion years in the Pilbara region of Western Australia, where the oldest known fossils on Earth are believed to exist.
First discovered in 1993, these rocks were thought to be fossilized remains of microorganisms similar to cyanobacteria, which were the earliest living organisms to produce oxygen on Earth. If confirmed, the presence of bacteria so early in our planet’s history would mean that Earth was predisposed to life much earlier than previously thought.
However, these findings have remained controversial. Researchers have repeatedly emphasized that the evidence could also be due to purely geological processes unrelated to ancient life. But through continuous AI training, we may one day further refine this innovative research technique.