The Future of AI

Engineers and computer scientists spent decades perfecting computers' abilities to solve classical math and logic problems. But as it would turn out, a huge set of real-world decision-making isn't readily framed as a tidy math problem. Machine learning (ML) earns its paycheck in these kinds of situations: When we're unable to logically or cost-effectively use math to tell a computer what to do, we can use ML to teach a computer what to do by showing it examples of how it's been done. This current AI/ML "Cambrian explosion" is resulting in a radical rethink of what computers can realistically learn. Startups and incumbents alike are teaching machines to emulate an ever-increasing share of capabilities once thought of as "uniquely human." Other frontiers of AI advancement include sensation and discernment (the five "senses"); creativity (reading, writing, and the arts); and congeniality (emotional intelligence).