Powerful AI finds 100 hidden planets in NASA data including extreme worlds (www.sciencedaily.com)

Astronomers at the University of Warwick have employed a cutting-edge AI tool named RAVEN to analyze data from NASA’s TESS mission, leading to the confirmation of over 100 exoplanets, including 31 newly identified worlds. This breakthrough is significant as it enables the discovery of rare and extreme planets, such as ultra-short-period planets that complete orbits in under 24 hours and those located in the "Neptunian desert," a region with previously thought to be scarce planets. RAVEN’s ability to filter out false signals from eclipsing binary stars enhances the reliability of these findings, making it a vital asset in modern astronomy. The findings, published in the Monthly Notices of the Royal Astronomical Society, are based on data from over 2.2 million stars and represent one of the most precise catalogs of close-in planets to date. RAVEN's comprehensive process—from signal detection to statistical validation—allows for a deeper understanding of planetary prevalence around Sun-like stars, revealing that approximately 9-10% host close-in planets, significantly refining previous estimates from the Kepler mission. By improving the methodologies for planet detection and providing interactive tools for further exploration, RAVEN is paving the way for a new era in planetary science, underscoring the transformative impact of AI in astrophysics.