Artificial intelligence tools are helping citizen scientists discover undiscovered exoplanets (artist illustration). Credit: Hypersphere/SPL
In September, artificial-intelligence expert Pavel Rabtsevich asked an AI model to search telescope data for previously undetected planets. About two weeks later, the system identified signals that could come from two planets orbiting a star 116 light-years from Earth. Professional astronomers have independently examined the signals, which are now listed as planet candidates, although the findings have not yet been peer-reviewed.
Rabtsevich, who works in Spain and has no formal astronomy training, is part of a growing group of amateur astronomers using AI to investigate worlds beyond the Solar System. Citizen scientists have been discovering planets for decades with tools published by NASA and other researchers. But AI models can now locate those tools online and apply them to large datasets, making it easier for non-experts to conduct searches.
Michelle Kunimoto, an exoplanet researcher at the University of British Columbia in Vancouver, Canada, cautions that not all amateur researchers are using these capabilities responsibly. Some planet hunters do not fully understand the algorithms used by AI systems, making it difficult for them to validate the results. Kunimoto says there has been “an overwhelming surge in AI-driven papers” and that “the systems we use to disseminate science are being overwhelmed.”
An AI-assisted search
Rabtsevich used Claude Code, an AI coding tool developed by Anthropic, for his search. He directed it to analyze observations of more than 120,000 stars made by NASA’s Transiting Exoplanet Survey Satellite (TESS) in late 2025 and early 2026.
He instructed Claude Code to apply a widely used planet-detection algorithm that searches for dips in a star’s brightness. Repeated dimming at regular intervals can indicate that a planet is orbiting the star and blocking a small amount of its light each time it passes between the star and the observer.
The system focused on a star designated TIC 4206066. Its data showed two distinct dimming cycles: one occurring approximately every three days and another approximately every 11 days. The pattern suggests that two planets may orbit the star, with the planet producing the three-day signal orbiting closer to the star.
Rabtsevich then asked Claude Code to search archived TESS observations for the same repeating patterns. The data collected in 2018 and 2020 contained the same dimming signals from TIC 4206066. At Rabtsevich’s direction, Claude Code also located a planet-detection algorithm created and published by Kunimoto, which Rabtsevich used to verify the signals.
Rabtsevich reported the findings in a preprint posted to Zenodo on September 25.1 The work has not undergone peer review, but TESS astronomers carried out an independent analysis and added the two objects to the mission’s official list of planet candidates.
Source: www.nature.com


