AI spots huge asteroid the size of 182 beavers that humans missed

An AI algorithm was able to spot a massive potentially hazardous asteroid, 2022 SF289, that humans had missed, showing the capabilities of this technology for planetary defense.

 An illustrative image of an asteroid passing Earth. (photo credit: INGIMAGE)
An illustrative image of an asteroid passing Earth.
(photo credit: INGIMAGE)

An artificial intelligence-powered algorithm has managed to discover a huge potentially hazardous asteroid the size of roughly 183 North American beavers that passed the Earth in 2022 – but scientists never even noticed.

The asteroid in question was designated 2022 SF289, because even though it was only recently discovered, the flyby still happened in 2022.

Though the asteroid poses no threat to Earth any time soon, the fact that the algorithm – known as HelioLinc3D and part of the Chile-based Vera C. Rubin Observatory – can discover it shows that these sorts of algorithms have the potential to better spot any potentially dangerous asteroids that orbit the Earth.

Leave it to algorithm: How big was the asteroid that passed the Earth?

According to a press statement from the University of Washington, asteroid 2022 SF289 is roughly 182.8 meters in diameter. But to put that in perspective, let's consider the size of a humble rodent, albeit the largest rodent in North America: The beaver.

According to the National Wildlife Federation, the North American beaver, the great national symbol of Canada, is around 90 centimeters long, not including its 35-centimeter-long tail.

 An illustrative image of a beaver, the national animal of Canada, which is 1/182th the size of asteroid 2022 SF289. (credit: PXFUEL)
An illustrative image of a beaver, the national animal of Canada, which is 1/182th the size of asteroid 2022 SF289. (credit: PXFUEL)

So, doing the math, one can find that asteroid 2022 SF289 is around 182 meters in size, assuming the tail is factored in.

Asteroid 2022 SF289 is also large enough to cause considerable damage to the Earth if it impacted, with the University of Washington's Institute for Data Intensive Research in Astrophysics & Cosmology (DiRAC) saying in a video that if it hit the Earth, it would have destroyed several cities.

Oh, dam: Can we stop an asteroid impact?

Humanity has developed a way to stop an asteroid impact, and it isn't through an extremely creative use of beaver dams. Rather, it is through finding a way to stop the asteroid from hitting the planet.

The most promising planetary defense method thus far was showcased in NASA's Double Asteroid Redirection Test (DART) Mission, which successfully changed an asteroid's orbit when tested.

In other words, humanity does have the means of stopping an asteroid impact. But that supposes we have enough warning to be able to pull it off, and that's not always something we have.

So an AI found an asteroid: Why does it matter?

Having a way to stop an asteroid impact means nothing if you don't know an asteroid is even there until it's too late. That is why scientists have also invested heavily into telescope systems to better spot them, and it's why this AI algorithm's success in finding one that humans missed is so significant.

There are millions of asteroids in the Solar System, with around 30,000 of them being designated Near Earth Objects (NEOs). Of these, around 2,350 of them have been designated potentially hazerdous, meaning they are at least 140 meters in size and within 7.5 million kilometers from the Earth. 

Scientists keep a close eye on these asteroids and try to identify as many as possible, as well as small ones that could still impact Earth, but plenty still fall through the metaphorical cracks, with several near-misses and even some asteroid impacts having happened from asteroids scientists didn't see coming.

Why asteroids can get missed varies on a case-by-case basis. In the case of 2022 SF289 and many others, it is because bright light sources obscured them from detection.

However, the AI algorithm HelioLinc3D was able to detect it by reviewing old data taken by the NASA-funded Asteroid Terrestrial-impact Last Alert System (ATLAS) in Hawaii. Going over it, the algorithm was able to identify the asteroid nobody else was able to.

And if it could do this, then it can possibly do it again for other asteroids that scientists have yet to find.

“This is just a small taste of what to expect with the Rubin Observatory in less than two years, when HelioLinc3D will be discovering an object like this every night,” DiRAC Institute and Rubin scientist Mario Jurić, leader of the team behind HelioLinc3D, said in a statement. 

“But more broadly, it’s a preview of the coming era of data-intensive astronomy. From HelioLinc3D to AI-assisted codes, the next decade of discovery will be a story of advancement in algorithms as much as in new, large, telescopes.”

Rubin is set to become fully operational in 2025.