Prehistoric predator? Artificial intelligence says no

Artificial intelligence has revealed that prehistoric footprints believed to have been made by a vicious dinosaur predator actually came from a shy herbivore.

In an international collaboration, University of Queensland paleontologist Dr Anthony Romilio used AI pattern recognition to reanalyze footprints from Dinosaur Stampede National Monument, southwest of Winton, central Queensland.

“Large dinosaur footprints were first discovered in the 1970s at a track site called Dinosaur Stampede National Monument, and for many years were believed to have been left by a predatory dinosaur, such as australovenatorwith legs almost two meters long,” said Dr Romilio.

“The mysterious traces were thought to have been left in the mid-Cretaceous, around 93 million years ago.

“But figuring out which dinosaur species exactly created the footprints – especially tens of millions of years ago – can be quite a difficult and confusing undertaking.

“Especially since these large tracks are surrounded by thousands of tiny dinosaur footprints, leading many to believe that this predatory beast could have unleashed a stampede of smaller dinosaurs.

“So to solve the problem, we decided to use an AI program called Deep Convolutional Neural Networks.”

It was formed with 1,500 dinosaur footprints, all of either theropod or ornithopod origin – the relevant dinosaur groups for the Dinosaur Stampede National Monument footprints.

The results were clear: the tracks had been made by a herbivorous ornithopod dinosaur.

Lead author Dr Jens Lallensack from Liverpool John Moores University in the UK said IT assistance was vital as the team was initially at an impasse.

“We were pretty stuck, so thank goodness for modern technology,” Dr Lallensack said.

“In our three-person research team, one person was pro-meat eater, one person was undecided, and one was pro-plant eater.

“So to really check our science, we decided to seek clarification from five experts, in addition to using AI.

“The AI ​​was the big winner, overwhelmingly outperforming all the experts, with a margin of error of around 11%.

“When we used AI on the large tracks at Dinosaur Stampede National Monument, all but one of those tracks were confidently classified as left by an ornithopod dinosaur, our prehistoric ‘predator’.”

The team hopes to continue enriching the database of fossil dinosaur tracks and conduct further AI investigations.

The research is published in the Journal of the Royal Society Interface and includes collaborations between Australian, German and UK researchers.

A replica of the Dinosaur Track is on display at the Queensland Museum in Brisbane, and the track site can be visited near South West Winton, Queensland.