Unless aliens are living somewhere nearby in the solar system, it’s unlikely that scientists will find direct evidence of extraterrestrial entities. But that doesn’t mean it’s impossible to find evidence of life elsewhere in the universe. Astrobiologists look for signatures of life, typically in the form of telltale biological chemicals like molecular oxygen and ozone in an exoplanet atmosphere.
However, even if astrobiologists found an exoplanet atmosphere containing these chemicals, it’s possible that some unknown or obscure process made them in the absence of life. If scientists found evidence of technological activity in space, or a technosignature, that would be more conclusive evidence of intelligent life elsewhere in the universe. In 1984, researchers founded the independent research institute known as the Search for Extraterrestrial Intelligence (SETI), dedicated to searching specifically for technosignatures in the form of radio signals.
From 2006 to 2020, the SETI@home project worked alongside researchers who were surveying the sky for excess radio emissions from space for their own projects using the Arecibo Telescope, which collapsed in December 2020. The SETI@home team gathered 400 days of combined observation time over these 14 years, producing billions of excess radio wave detections from different points in space. However, they think the vast majority of these won’t be useful for finding aliens, since they come from radio frequency interference, benign sources like pulsars or gas clouds, or scattered points in the sky rather than a single source.
To clean up their radio emission data, the team recently developed algorithms that remove interference and identify signals from fixed sources. This strategy prepared the researchers for their final step: re-observing these locations with the Five-hundred-meter Aperture Spherical Telescope, or FAST.
The radio signals that these algorithms are designed to detect would have to appear distinct from signals produced by natural sources in space. Therefore, the team established 3 criteria for detectable technological signals. First, they had to be fixed within a narrow frequency range over an extended period. Second, they had to pulse at a constant cycle. Or third, they had to contain repeating structures lasting several seconds.
One caveat they addressed is that intentionally transmitted signals intended for detection would likely differ from radio waves that leaked from an alien atmosphere and were observed coincidentally. An intelligence trying to communicate would account for how the motion of its planet around its star would affect the outgoing frequency of its signal, a phenomenon known as the Doppler Shift. To account for this issue, the team reasoned that aliens trying to communicate would produce radio signals at a nearly constant frequency, which would be a clear giveaway of an artificial radio source.
During the development and testing of their algorithm, the researchers also included artificial data points intended to simulate hypothetical detections of definite technosignature radio signals. They called these artificial data points candidate birdies. If their algorithm identified the birdies as targets for further investigation, then they could be confident that it was working properly. They adjusted the algorithm’s detection sensitivity and filtering based on whether it included or excluded the birdies from the targets for further investigation.
The team addressed the challenges of filtering and scoring their data by breaking them up into smaller tasks that could run on multiple machines simultaneously. For a run of their algorithm on 2,000 connected computer processors, filtering took around 15 hours, while scoring took 1.6 days. They completed 2 runs of their algorithm on the SETI@home data, including 1 run with 3,000 birdies for comparison. The team used the birdies to determine which settings of their algorithm yielded the largest number of targets for further investigation above certain energy thresholds and across different radio frequencies. By doing so, they identified 92 candidate signals of interest as priorities to re-observe using 23 hours of observation time they secured with FAST.
The work of re-observing and analyzing these signals is ongoing. As of July, 2025, the researchers had re-observed 80 of the 92 candidate signals. They haven’t found any evidence of extraterrestrial intelligence yet, but they suggested that dedicated radio telescope surveys would help. In the meantime, the expense and demand for radio telescope time mean that SETI can still likely gather the most data by partnering with other radio astronomers, gaining access only to what they observe.
