We analyze per-second GPS driving data to create a unique DriverDNA for each trip. We compare this DriverDNA against our database of over 7 billion driving patterns, trained with damage data, saved CO2 emissions and fuel consumption from real trips.
We’ve been training our AI since 2004, using data from over 106 countries and 1,600 cities. This extensive knowledge enables us to rapidly predict crash probability and climate impact.
You might think of our pattern AI as being similar to facial recognition. However, instead of comparing faces to a database of known faces, we compare driving patterns to a database of known trips and their outcomes.