AI finds 100+ new planets and 2,000 possible new worlds using NASA's TESS data

In a development that signals a quiet but significant shift in space research, scientists have used artificial intelligence to validate over 100 planets beyond our solar system – while flagging more than 2,000 additional signals that could point to new worlds.

The findings, led by researchers from the University of Warwick and published in the journal Monthly Notices of the Royal Astronomical Society, show how automated systems are beginning to take on tasks that once required months of manual effort by astronomers.

A system designed to catch what humans might miss

At the center of the research is an AI-based pipeline called RAVEN. Instead of relying on scientists to manually scan through vast amounts of telescope data, the system processes it independently – identifying patterns that are often too subtle or time-consuming to detect.

The data was sourced from Transiting Exoplanet Survey Satellite (TESS), a space telescope that continuously monitors stars and tracks tiny dips in their brightness.

How planets reveal themselves

These small dips are key.

When a planet passes in front of a star, even briefly, it blocks a fraction of the star’s light – making it appear slightly dimmer. This technique, known as the Transit Method, has been one of the most reliable ways to detect planets outside our solar system.

Using this method, the AI system identified:

Over 100 validated planets, including dozens newly detected

More than 2,000 additional signals that could potentially be planets

Why this marks a shift in space research

Until recently, analyzing such signals depended heavily on human effort – astronomers manually reviewing light curves, verifying patterns, and ruling out false positives.

But the scale of data has grown dramatically.

Tools like RAVEN are now helping scientists filter through this information faster, narrowing down the most promising candidates in a fraction of the time. Rather than replacing astronomers, the technology is reshaping how they work – allowing them to focus on confirmation and deeper analysis.
Thumb image: Canva (for representative purposes only)