3 Reasons Technosignatures Detected by AI-Trained Algorithm Can Be Extraterrestrial Activities

Astronomers picked up extraterrestrial signals which they previously missed in an area they thought was devoid of potential ET activity. It could be the first hint that humans are not alone in the universe.

Mysterious Signals Detected

Experts led by University of Toronto student Peter Ma used an algorithm with artificial intelligence (AI) to examine 820 stars in an area they didn't suspect would have any potential activity. They were surprised with their finding, especially since they missed the tentative signals earlier due to a lot of interference, Daily Mail reported.

Ma, along with astronomers from the SETI Institute, Breakthrough Listen and scientific research institutions around the world, developed a new machine-learning algorithm that can detect potential extraterrestrial signals from the background noise on our planet.

The algorithm included deep learning, a type of machine learning and AI that imitates how humans gain a certain type of knowledge. The same type of AI is behind Tesla's driverless cars. They taught the machine to differentiate human-caused noise from potential alien signals.

The machine did not detect any radio signals from the selection of the stars collected by the Robert C. Byrd Green Bank Telescope in West Virginia.

However, Ma and his colleagues detected eight signals coming from the area. According to Steve Croft, a project scientist for Breakthrough Listen on the Green Bank Telescope, the most important aspect of any technosignature search is sifting through this vast haystack of signals to find the needle that could be an alien transmission.

Most signals detected by their telescopes originate from their own technology, such as GPS satellites and mobile phones.

Peter's algorithm provides them with a more efficient way to filter the haystack and locate signals with the expected characteristics of technosignatures. The eight extraterrestrial signals detected from five of the 820 stars were located between 30 and 90 light-years away.

ALSO READ: Space Probes Between Humans, Extraterrestrials Will Have Extraordinary Disparity That We Couldn't Reverse-Engineer [REPORT]

3 Reasons Technosignatures Detected Could Be Signs of Extraterrestrial Activities

The researchers were convinced that what they detected was what extraterrestrial signals would look like for several reasons.

First, they were narrow bands, so they were not natural phenomena because the latter typically generate signals with a broad spectrum.

Second, they were also equipped with what is known as a"slope," indicating that the origin of the signals was accelerating relative to our antennas and therefore could not have originated from our planet.

Lastly, they appeared in ON-source observations as opposed to OFF-source observations. Human radio interference typically occurs in both ON- and OFF-source observations due to the proximity of the source.

Croft added that they are present when they look at the star and absent when they look away, unlike local interference which is typically always present.

Additionally, the changing frequency of the signals over time causes them to appear distant from the telescope.

They believe it will accelerate the rate at which they are able to make discoveries with their grand effort to answer the question, "Are we alone in the universe?"

Unfortunately, they found one issue — after subsequent observations, they have not detected the radio signals again.

Astronomers are eager to examine them in greater detail and determine if they are truly from deep space or merely terrestrial interference. So, they have to detect them again.

Nonetheless, the new algorithm has a great deal of potential for future searches for extraterrestrial civilizations.

Cherry Ng, one of Ma's research advisors, said the dramatic results demonstrate the power of applying modern machine learning and computer vision methods to data challenges in astronomy, resulting in both new detections and increased performance.

The application of these techniques on a large scale will revolutionize radio technosignature science.

The study was published in the journal Nature Astronomy.