AI could help us discover aliens: there are ‘suspicious signals’

“Tell me about the existence of very distant worlds” recited a famous song by Battiato from the 1980s. The possibility of being able to pick up signals coming from possible alien civilizations has always been …

AI could help us discover aliens: there are 'suspicious signals'

“Tell me about the existence of very distant worlds” recited a famous song by Battiato from the 1980s. The possibility of being able to pick up signals coming from possible alien civilizations has always been one of humanity’s great dreams. Today we are trying this also thanks to the (significant) help of Artificial Intelligence. It is one of the bases of Seti’s “Cosmic” project, an acronym for “Search for Extraterrestrial Intelligence”. The organization, an active partner of NASA, is a research body active since 1984. Its mission? Understanding whether intelligent life forms exist beyond our planet.

How the “Cosmic” project works

“Cosmic” will use data from a series of satellite dishes, positioned in the New Mexico desert, to find extraterrestrial radio signals. It is a system of 27 radio telescopes called VLA (Very Large Array) capable of working as a single receiving antenna of 40 km in diameter. Their job is to observe quasars, pulsars, black holes, galaxies and everything else astronomers want to study. However, for the “Cosmic” project these radio telescopes will also have to look for “technosignatures” or “technosignatures” in the next two years.

Technosignatures are evidence of environmental alterations caused by technology that we humans can detect and which could constitute evidence of the existence of an extraterrestrial civilization. Examples of this type can be: radio signals or laser emissions.

What does artificial intelligence have to do with it?

In the case of “Cosmic” we are talking about radio signals with a frequency between 0.75 and 50 GHz which are then digitally processed in real time by the system. And this is where artificial intelligence could make a difference. Finding radio signals coming from other galaxies is not easy. First of all for what is called “cosmic background noise”, i.e. the random noise that is found outside the Earth’s atmosphere and which originates from the activities of celestial bodies in space. This “noise” is very high and makes it very difficult to distinguish signals coming from physical bodies from those coming from “alleged” alien civilisations.

Not only that: even radio waves coming from terrestrial activities, such as those from television, radio or satellite broadcasts, can “confuse” detection instruments. But the enormous data processing and self-learning capacity of new AI software could prove fundamental in these analyses.

“Two thousand suspicious signals per hour”

The collected data is analyzed in real time by algorithms based on machine learning models. Translated: the more data they analyze, the greater their ability for these systems to distinguish between all these inputs and detect signals that can then actually be analyzed by scientists. A sort of “mega-skimming” which however becomes essential, also due to the great abundance of variables to manage.

At the time, Seti officials had said that the system was capable of picking up signals at a distance of up to 81 light years. This means that, in the worst case scenario, the signal we managed to pick up is something like 800 thousand billion km away. It seems like a lot, but it’s a pretty close distance as far as the cosmos is concerned.

In reality there are not many details today on the distance that can be reached in the “harvesting”. What is certain is that it will be possible to capture even short signals, of the order of a few nanoseconds which, according to scientists, are those that a probable “alien” civilization could emit.

The results? So far, according to what the Spanish version of Wired reports, 500 thousand suspicious audio sources have already been scanned, at a rate of two thousand per hour. And it could just be the beginning.