Prehistoric predator? Artificial intelligence says no

Dinosaur footprints thought to be made by a fearsome predator were actually made by a timid animal.

The footprints from the Dinosaur Stampede National Monument were re-analyzed with the help of artificial intelligence.

Dinosaur footprints were first discovered in the 1970s at a track site called the Dinosaur Stampede National Monument, and for many years they were believed to be left by a predatory dinosaur.

The tracks were thought to be from the mid-Cretaceous Period.

It can be hard to work out what species of dinosaurs made the footprints.

Since these big tracks are surrounded by thousands of tiny dinosaur footprints, many people think that this predatory beast could have caused a lot of smaller dinosaurs to flee.

To crack the case, we decided to use an artificial intelligence program.

It was trained with 1,500 dinosaur footprints, all of which were theropod or ornithopod in origin.

The tracks were made by a dinosaur.

The computer assistance was crucial as the team was at an impasse.

Thanks to modern technology, we were stuck.

One person was pro- meat-eater, one person was undecided, and one person was pro-plant-eater.

To really check our science, we went to five experts and used artificial intelligence.

The artificial intelligence was the clear winner, beating all of the experts by a wide margin.

All but one of the large tracks from the Dinosaur Stampede National Monument were classified as left by an ornithopod dinosaur by the use of the artificial intelligence.

Adding to the fossil dinosaur tracks database is one of the things the team hopes to do.

Australian, German, and UK researchers collaborated on the research which was published in the Journal of The Royal Society interface.

There is a replica of the dinosaur trackway on display at the Qld Museum, and it can be visited near Winton.

A machine learning approach for the discrimination of theropod and ornithischian dinosaur tracks was written by Jens N. Lallensack and her team. The DOI is 10.1098/rsif. 2022.0588.

Journal information: Journal of the Royal Society Interface