AI has revealed over 100 hidden exoplanets in NASA’s data, with thousands more yet to be confirmed. Seven years into its mission, TESS has already identified approximately 700 new worlds. However, not all signals flagged as promising have been properly sorted—many remain undetected. A research team recently validated nearly 1,000 of these worlds, including those from an unverified pipeline called RAVEN, which was developed to distinguish real planets from false positives. RAVEN works by using machine learning to analyze simulated planet signals and identify patterns that could indicate actual planetary systems. This process involves flagging potential candidates, validating their existence through mathematical analysis, and confirming the true nature of the objects detected. The pipeline was designed to handle a stubborn challenge: separating genuine planets from artificial signals that mimic their appearance. Dr. Marina Lafarga Magro, a postdoctoral researcher leading the project, explained how RAVEN transformed a large dataset into a catalog of verified planets. The results show that while 118 planets were successfully validated, many more high-confidence candidates remained unconfirmed. These findings suggest that the search for exoplanets is becoming more precise as tools evolve, especially with the help of advanced simulations like RAVEN. A deeper look reveals that some of the most intriguing discoveries involve two rare categories: ultra-short-period planets—worlds that complete orbits in less than 24 hours—and the mysterious Neptunian desert, where Neptune-sized planets are unusually absent. Another category includes tightly packed multi-planet systems with previously unknown companions. Notably, a recent study found that close-in planets occur about 9–10% of Sun-like stars, significantly reducing uncertainty compared to previous missions like Kepler. The same clean sample used in this study also provided the first head count of planets in the Neptunian desert, indicating a much lower prevalence of such worlds in solar-type stars. As scientists continue refining their methods, the next step is to analyze these new data points to better understand the formation and stability of planetary systems. With improved tools, researchers can now more accurately measure the rarity of these unusual worlds, opening up new avenues for exploration and scientific discovery.
