Could AI find alien life before us?
New machine-learning system could tell the origin of samples found on Earth and elsewhere

Scientists have developed an artificial intelligence (AI) tool in their quest to discover alien life.
A new machine-learning algorithm has been trained for this purpose, making use of living cells and fossils, alongside lab-made chemicals, to look for life on other planets.
The AI, developed by a team led by Robert Hazen, from Carnegie Science in Washington, and Jim Cleaves, of Tokyo Tech, can "distinguish between samples of biological and nonbiological origin 90% of the time", Live Science reported.
Subscribe to The Week
Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.

Sign up for The Week's Free Newsletters
From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.
From our morning news briefing to a weekly Good News Newsletter, get the best of The Week delivered directly to your inbox.
However, the "algorithm's inner workings remain a mystery", said the website, as AI systems are "largely black-box models – viewed only in terms of their inputs and outputs". As a result, researchers are still unsure about how the system reaches its conclusions.
Regardless, it could offer "important evidence" regarding the chemistry of life following "different fundamental rules than those of the nonliving world", the website added.
What did the papers say?
The team's findings were detailed in a paper published in the Proceedings of the National Academy of Sciences.
If successful, the new test could be used "almost immediately" and could search through rocks collected by the Mars rover, as well as ancient rocks on Earth, said Live Science.
Using a machine in this way is "different to techniques scientists have used in the past", said BBC Newsround, as it is not looking for specific things, but "differences between samples".
The AI utilised "pyrolysis gas, chromatography and mass spectroscopy" to look at the "subtle differences within a sample's molecular patterns", Cosmos Magazine explained. And it was able to identify living things including "shells, teeth, bones, leaves, rice, human hair and cells in rock".
In wider space exploration developments, some space agencies have planned missions to Mars, "with the aim of collecting and returning samples back to Earth", Euronews reported, including the Mars Sample Return Mission devised by Nasa and the European Space Agency.
Such missions could use the AI technology developed by this team of scientists in their search for answers.
Hazen explained that the technology "opens the way to using smart sensors on robotic spacecraft, landers and rovers to search for signs of life before the samples return to Earth".
What next?
As AI develops, this new technique could be used to "settle debates over the origin of a number of samples found on planet Earth", such as "3.5 billion-year-old sediments found in Western Australia", Euronews reported.
It is also being used in the Breakthrough Listen project at the University of California, Berkeley, searching the universe for signals from alien technology. It is trained to spot familiar signals from human technology and flag anything out of the ordinary.
For now, humans are still "intimately involved" in the process, as they have to "follow up and investigate" anything spotted by the AI, said Space.com.
The algorithms "aren't that smart" yet, but the time may soon come when they are. Artificial intelligence might be able to think of ways to search for alien life "beyond the confines of human biases and experience", such as finding ways to communicate through signals in space. In the quest for answers, said Space.com, AI has the "capability to do things faster and better than humans can".
The question is would it then translate all of that information back to humans. If AI did detect alien life, "we may not get the full picture".

Continue reading for free
We hope you're enjoying The Week's refreshingly open-minded journalism.
Subscribed to The Week? Register your account with the same email as your subscription.