Amateur astronomer harnesses AI to find a possible hidden exoplanet
Evidence for the possible super-Earth has been lurking for years in NASA data. Pavel Rabtsevich found it—with a little help from AI
Amateur astronomer Pavel Rabtsevich discovered a potential hidden exoplanet using artificial intelligence (AI) algorithms. Rabtsevich, a space-science hobbyist based in Spain, examined data collected by NASA’s Transiting Exoplanet Survey Satellite (TESS) for anomalies. The TESS telescope searches for the slight dimming of stars caused by planets transiting in front of them.
Rabtsevich's AI-assisted analysis zeroed in on TIC 4206066, a star located about 116 light-years away in the Orion constellation. The data pointed to dimming intervals of around two hours every three days. After further investigation, Rabtsevich found 22 additional instances of this dimming pattern. If confirmed, the exoplanet would be a super-Earth approximately 1.4 times the size of Earth.
However, its proximity to TESS's thruster firings may have introduced some uncertainty. Rabtsevich acknowledges that the signal could be due to other factors and emphasizes the need for independent verification. Despite the potential for another exoplanet, the discovery was made possible by AI collaboration, with Rabtsevich utilizing multiple AI agents to write and verify code.
Other astronomers, including Pablo Mercader Perez, have corroborated Rabtsevich's findings using different methods and software. NASA's TESS team is now reexamining the star in hopes of confirming the exoplanet's existence.
Written by urgent.news from Scientific American's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
