A group leading the search for extraterrestrial life discovered some unusual radio signals in space. Peter Ma, a student at the University of Toronto, described this incredible find as a "very suspicious" signal that was detected by a new artificial intelligence (AI) system.
The new machine-learning algorithm was created and trained to detect such radio signals. It was brought to the Australian-based program called Breakthrough Listen, which is breaking new ground in the search for extraterrestrial intelligence (SETI).
New AI Trained to Look for Extraterrestrial Life Detects 8 'Very Suspicious' Signals: Are Those Worthy of Study?
Using AI to Detect Extraterrestrial Life
Scientists believe that the world may be witnessing a Cambrian Explosion in AI with examples like the Midjourney, DALL-E 2, and ChatGPT, which demonstrated the quick advance humans have achieved in machine learning.
Researchers noted in their article in The Conversation that AI is currently employed in almost every field of science to assist researchers with mundane categorization tasks. More so, it is assisting the team of radio astronomers in broadening the hunt for extraterrestrial life, with encouraging results thus far.
As scientists search for evidence of intelligent life outside Earth, they built an AI system that beats classical algorithms in detecting radio signals from space. The novel AI system was trained to search through millions of data from radio telescopes for signals that are out of this world.
Upon feeding the AI system with previously studied datasets, they found bizarre eight signals that classical algorithms missed. But they also pointed out the possibility that these very suspicious signals may not come from extraterrestrial intelligence but more likely a case of radio interference.
Nonetheless, their findings in the study, titled "A Deep-learning Search for Technosignatures From 820 Nearby Stars" published in the journal Nature Astronomy, highlighted how AI techniques significantly contributed to the search for extraterrestrial life.
Why AI Is Not That Intelligent
But the team clarified that AI algorithms are not that smart because the system does not understand or think. Rather, it excels at pattern recognition, which has been proven exceedingly useful in the field of classification, and only does tasks they were trained at doing,
Radio astronomers use AI to look for radio "technosignatures," which are hypothesized to indicate the presence of technology and by proxy mean the existence of a society capable of communication. The AI system they created classified the signals as either radio interference or genuine technosignatures.
According to New Atlas, the AI identified 20,515 signals of interest out of the 3 million signals in the dataset. Researchers inspected each one of these manually and found that eight of these signals have characteristics of being technosignatures, which could not be linked to radio interference.
Ma noted that these eight signals were deemed worthy to study after carefully examining them. However, they did not see the source of the signals again when they took another look a the targets using telescopes. It has been five to six years since they collected the data and has not received any signal again.
But as intriguing as these signals might be, researchers said that it is still far from an answer of whether extraterrestrial life exists. They acknowledge that they do not know what a real extraterrestrial technosignature would look like and that they might be using AI to look for the wrong things. Still, it was a worthwhile experience applying AI to yield more technosignatures.
RELATED ARTICLE: Has Extraterrestrial Life Finally Got into Contact? Astronaut Spots Stowaways Outside Cargo Spacecraft Resupplying the International Space Station
Check out more news and information on Space in Science Times.
