A new AI-powered app called DinoTracker is revolutionizing the way scientists and the public identify ancient dinosaur footprints. Developed by a research team led by the Helmholtz-Zentrum in Berlin and the University of Edinburgh, the app uses artificial intelligence to analyze photos or drawings of fossil tracks and instantly estimate which type of dinosaur likely made them. For decades, paleontologists have struggled to interpret fossil footprints due to their distorted shapes caused by geological processes over millions of years. Traditional methods relied on manual databases linking tracks to specific species, which could introduce human bias—especially when identifying controversial or poorly understood prints. To overcome this, the team trained an AI system on nearly 2,000 real fossil footprints and millions of simulated variations that mimic natural changes like compression and edge deformation. The model learned to recognize eight key features, including toe spread, heel position, contact area, and weight distribution. By comparing these traits, the AI can predict the most likely dinosaur species with about 90% accuracy—matching expert assessments in most cases. One of the most surprising outcomes was the discovery of striking similarities between some 200-million-year-old footprints and those of both ancient and modern birds. This raises the possibility that birds may have evolved much earlier than previously thought, or that certain dinosaurs had foot structures resembling those of birds by chance. The app also shed light on mysterious tracks found on the Isle of Skye in Scotland, dating back 170 million years. The AI analysis suggests they were made by some of the earliest known relatives of duck-billed dinosaurs—among the oldest examples of this group ever found. Beyond research, DinoTracker opens paleontology to the public, allowing anyone with a smartphone to contribute to fossil studies. The technology offers an objective, data-driven approach to classifying footprints, reducing reliance on subjective judgment. The study, published in PNAS, was supported by the BMBF-Project Data-X, the Helmholtz project ROCK-IT, the Helmholtz-AI project NorMImag, the National Geographic Society, and the Leverhulme Trust. Dr. Gregor Hartmann of Helmholtz-Zentrum described the tool as a powerful, unbiased method for analyzing footprint variation and testing scientific hypotheses. Professor Steve Brusatte of the University of Edinburgh called it a groundbreaking advancement in paleontology—one that could help uncover when major groups like birds first appeared, and how dinosaurs moved across ancient landscapes.