Finding planets that could potentially be home to life isn’t very easy, but researchers from Switzerland have created an AI model that has the potential to make it much more practical. The team wrote a paper, which has been published in the journal Astronomy and Astrophysics, about the new tool.

The new algorithm has already been used to spot 44 star systems that it believes contain exoplanets that are similar enough to Earth that they have the potential to support life. These are all exoplanets that had not previously been detected. The astronomers need to go back and study each one to see if the planets actually can (or maybe even do) support life.

This new tool is primarily designed to help point astronomers in the right direction in their search for life-supporting planets, and if simulations are correct, it is extremely good at its job.

Researchers ran a number of different simulations and found that the AI was able to identify systems where these planets would be 99% of the time. Dr. Yann Alibert is the co-director of the University of Bern’s Center for Space and Habitability and the co-author of the study. Forbes quoted him as saying:

“It’s one of the few models worldwide with this level of complexity and depth, enabling predictive studies like ours. This is a significant step in the search for planets with conditions favorable to life and, ultimately, for the search for life in the universe.”

The fact that the AI has pointed out 44 star systems that warrant further investigation is amazing on its own. Scientists have only identified 5800 confirmed planets outside of our own solar system, and that isn’t just ones that may be able to support life. Having an AI that can help direct astronomers and other researchers in the right direction is invaluable.

The AI was trained by feeding it synthetic planetary systems that were generated using the Bern Model of Planet Formation and Evolution. This is a model that simulates planetary development back in time all the way to their creation as protoplanetary discs. In a statement, Alibert said:

“The Bern Model is one of the only models worldwide that offers such a wealth of interrelated physical processes and enables a study like the current one to be carried out.”

The team also fed in information from about 1600 systems that astronomers know have at least one planet and a star that is G-type, K-type, or M-type. Once all that data had been fed into it, the AI generated the 44 systems that it thought likely had a life-supporting planet.

The researchers admit that this system is not perfect, but it should only improve with time and more information. Plus, perfection isn’t necessary in this type of thing. Without the AI tool, the astronomers have to analyze every solar system to determine if it has planets, and then research more to see if any of the planets could support life. Given the fact that there are trillions of solar systems out there, this would be a very lengthy process.

Having an AI narrow that search is invaluable.

If you thought that was interesting, you might like to read about a quantum computer simulation that has “reversed time” and physics may never be the same.