Scientists may have detected more than 10,000 substances never seen before exoplanet A single investigation can quickly triple the number of known otherworlds. This record-breaking capture was made possible thanks to a new algorithm that helped researchers analyze more than 80 million stars, revealing subtle clues that would otherwise be “impossible” for us to see.
Since then, The first extraterrestrial planet was discovered in 1995the number of exoplanet discoveries is slowly increasing in line with new technologies such as: james webb space telescopehas great functionality for finding these strange other world. In September 2025, astronomers expect the number of confirmed exoplanets to reach There were over 6,000According to , nearly 300 have been added to the list since then. NASA.
The research team used machine learning algorithms to analyze the light curves of exactly 83,717,159 stars captured by NASA’s Exoplanet Survey Satellite (TESS), which has been orbiting Earth since 2018. TESS is a car-sized space telescope that has been orbiting Earth since 2018. Astronomers can tell when a planet has passed, or is likely to have passed, in front of its host star by looking for subtle drops in a star’s brightness.
This revealed more than 11,000 exoplanet candidates, 10,052 of which had never been seen before. (Other scientists had previously identified the remaining planets, but none have yet been confirmed as exoplanets.) About 87% of the candidates were found to make more than one pass, allowing the researchers to calculate the planets’ orbital periods ranging from 0.5 days to 27 days. Stella Catalog.com.
TESS is designed to look for objects that pass in front of distant stars. This wide-field image was one of the first taken shortly after launch in 2018.
(Image credit: NASA/MIT/TESS)
But the researchers didn’t stop there. To test the validity of their model, they sought to confirm one of the new candidates themselves.
The research team used one of the 6.5-meter-tall Magellan telescopes in Chile’s Atacama Desert to identify the exoplanet, dubbed a “hot Jupiter.” TIC 183374187 borbits a star about 3,950 light-years from Earth. Exactly where the algorithm predicted.
The confirmation of TIC 183374187 b suggests that at least some of the other exoplanet candidates will eventually be confirmed as well. However, first these planets need to be verified and studied in more detail by independent surveys, which could take months or years to do properly.
Find an “impossible” planet
TESS is specifically designed to detect transiting objects and has already discovered 882 confirmed exoplanets (about 14% of the current total), so it may seem strange that no one has seen most of the new candidates so far. However, there’s a good reason for that.
Most researchers prefer analyzing the light curves of the brightest stars in the TESS dataset. Because these star transit events are more pronounced and easier to see. But there are many more faint stars that will be captured in the telescope’s wide-field images.
In the new study, the researchers observed all the stars from TESS’s first wide-field images up to 16 magnitudes fainter than the usual standard for transit studies. The researchers believe this idea is T16 project.
The machine learning algorithms used in the new study looked for subtle fluctuations in a faint star’s light curve that could be caused by a planet “passing” an alien sun.
(Image credit: NASA/JPL)
These light curves are very faint, making potential transit events very difficult to spot, and therefore usually overlooked. To overcome this hurdle, the team created a machine learning algorithm that learns to distinguish between subtle clues that indicate a possible traffic stop. (Machine learning is a subset of) artificial intelligence Computers are not explicitly programmed to make predictions by learning from data. )
The team was also able to use computer programs to analyze huge data sets that would have been “impossible” for humans to classify on their own. Universe Today reported.
“This study shows that large-scale transit searches using machine learning can significantly expand the search for transit planet candidates, especially around faint stars,” the researchers wrote in their paper.
Unfortunately, the short orbital periods of exoplanet candidates mean that they are likely Too close to their home planet to support life as we know it. (This is because more distant planets orbit their stars less frequently and are less likely to coincide with observers for a transit.)
Roth, J. T., Hartman, J. D., Bakos, G. Á., Yee, S. W., Bouma, L. G., Galarza, J. Y., Teske, J. K., Butler, RP, Crane, J. D., Shectman, S., Osip, D., Vissapragada, S., Beletsky, Y., Kanodia, S., and Gaibor, Y. (2026). T16 Planet Hunt: 10,000 new planet candidates from TESS Cycle 1 and confirmation of hot Jupiter around TIC 183374187*. Astrophysical Journal Supplement Series, 284(1), 19. https://doi.org/10.3847/1538-4365/ae5b6c
