AI analyzed Hubble data, finding 1,300 unusual cosmic objects in days.

Astronomers using an AI tool called AnomalyMatch have discovered over 1,300 unusual cosmic objects in Hubble Space Telescope data, including more than 800 never before documented. The AI analyzed nearly 100 million image cutouts from the Hubble Legacy Archive, identifying rare phenomena such as merging galaxies, gravitational lenses, jellyfish galaxies with gas tails, and unclassifiable structures. The system, trained to detect anomalies through machine learning, completed the scan in just two and a half days—far faster than human review. The findings, published in Astronomy & Astrophysics, show how AI can uncover hidden insights in vast scientific datasets, paving the way for similar discoveries in data from upcoming telescopes.