AI Identifies More Than 1,300 Unusual Objects in Hubble Space Telescope Images
Artificial intelligence has combed through decades of Hubble Space Telescope data, revealing hundreds of unusual celestial objects that astronomers had missed. ESA’s AnomalyMatch neural network scanned around 100 million image cutouts in just days, flagging more than 1,300 anomalies. Among them are colliding galaxies, rare gravitational lenses, and jellyfish galaxies, demonstrating how AI can rapidly unlock discoveries hidden within vast astronomical archives. The project highlights a new era where machines partner with scientists to explore the universe efficiently and faster.
