Isabel Angelo spends her days searching the cosmos for exoplanets and signs of extraterrestrial intelligence. But Angelo’s head isn’t in the stars. It’s buried under mountains of 0s and 1s — data from space telescopes old and new.
To piece together a fuller picture of what’s out there (alien megastructures around nearby stars, perhaps?), Angelo and her colleagues at the SETI Institute in Mountain View, Calif., are turning to machine learning models. These tools are analyzing the troves of telescope data for anomalies — things that look “weird” in space that we don’t know how to explain yet.
Most of the space weirdos will probably turn out to be total normies — a comet, a newborn star or a star surrounded by lots of gas and dust. “These are things that I would expect [the models] to find,” Angelo says. “But the goal is to find something that’s not even in any of these categories.”
In the meantime, Angelo keeps her whimsy and wonder for space alive by remembering historic moments, like the reveal of the first portrait of a black hole in 2019. The scientists “put all of these telescopes all over the Earth. They wait for the Earth to rotate. They do all this math to construct this one image,” Angelo says. “I can’t believe how creative and innovative we have to be to learn about space. It’s such a specific and really fun challenge.”


