On the inside, we use artificial intelligence that allows the car to be powered by deep learning. We wanted to bypass the need to hard-code detection of specific features—such as lane markings, guardrails, bicyclists—and avoid creating a near-infinite number of “if, then, else” statements. That's too impractical to code when trying to account for the randomness that occurs on the road. This sort of “deep driving” can identify objects and intent, and can process piles of data. We’re using it for everything from building maps to identifying objects to combining the input from sensors. Deep learning is also used to offer a smoother ride by learning from examples. This eliminates jerkiness for a more natural feel.