Artificial intelligence is not a new concept. Machine learning has been used for decades in large server and high performance computing environments. Why is it attracting so much attention, investment and innovation now? The answer, as with most disruptive technologies, is a convergence of technology enablers and market demands that previously were insufficient to drive scale. In this presentation, we examine the market for processors in AI applications. We illuminate key trends in adoption of different classes of processors in a range of key applications, and explain why embedded vision is at the heart of these trends. And, we show how this adoption is driving revenue for different classes of processors in markets including automotive, consumer, data processing, industrial and wired and wireless communications.