AI finds hundreds of never-before-seen ‘cosmic anomalies’ in old Hubble Telescope images
When you purchase through links on our articles, Future and its syndication partners may earn a commission.
Credits: ESA/Hubble & NASA, D. O’Ryan, P. Gómez (European Space Agency), M. Zamani (ESA/Hubble)
The Hubble Space Telescope takes many photos. In fact, NASA estimates that Hubble has taken 1.7 million images since its launch in 1990. But that poses a unique problem: It’s almost impossible for scientists to examine all the images.
With this in mind, two researchers from European Space Agency (ESA) built an AI model called AnomalyMatch to search the vast Hubble Telescope dataset, and the AI managed to discover 1,300 anomalies, or strange-looking objects. Hundreds of these anomalies have never been documented before.
“This is a powerful demonstration of how AI can improve scientific feedback from archival datasets,” Pablo Gómez, one of the ESA researchers who built the model, said in a statement. statement.
Many of these remarkable new objects discovered actually defy classification, says NASA. Most were distant galaxies in flux as they merge and interact in strange ways and scientists specifically point to “galaxies with massive star-forming clusters, jellyfish-like galaxies with gaseous ‘tentacles,’ and planet-forming disks on the edge in our own hamburger-like galaxy.”
A lack of time
The images collected by Hubble represent the largest volume of observational data we can analyze in the history of astronomy, but this dizzying amount of information poses a barrier to scrutiny by human observers. There just isn’t enough time. That’s why it’s promising for NASA to say that it took the team less than three days to sift through nearly 100 million image cutouts using AnomalyMatch.
As for how it works? The researchers trained the AI model to detect strange objects using pattern recognition. AnomalyMatch was essentially designed to analyze images in the same way we process visual information in our brains.
NASA calls this project a significant step forward. This is the first time that a systematic search for astrophysical anomalies has been conducted across the entire Hubble Legacy archive, which spans decades of deep space observation.
“Archival observations from the Hubble Space Telescope now go back 35 years, providing a treasure trove of data in which astrophysical anomalies could be found,” David O’Ryan, lead author of the research paper, said in another article. statement.
“The discovery of so many previously undocumented anomalies in Hubble data highlights the potential of this tool for future investigations,” Gómez said.
Publication of Astronomy and Astrophysics the paper retailer AnomalyMatch and its results in December 2025.
