The father of modern AI studied and conducted research on the topic of artificial intelligence and robotics at the Technical University of Munich. Today, he is co-director at the Dalle Molle Institute for Artificial Intelligence Research in Lugano, Switzerland. Together with Sepp Hochreiter, a student of Schmidhuber’s at the time, he co-authored a paper as far back as 1997 that laid the groundwork for contemporary machine learning. The concept of long short-term memory is key to software solving difficult problems faster, better and more effectively by using recall—much as people do. Today, this technology can be found in such applications as Google’s speech recognition.
Schmidhuber takes the optimistic—and hence by no means discipline-wide—view that machines and software will soon develop a life and consciousness of their own, not unlike ours.
“As a child, I wanted to be a physicist because that involves researching the fundamental nature of the world. When I reached my teenage years, I realized that there is something even more important—to grasp the nature of understanding,” related Schmidhuber in a recent interview. “I thought to myself: I should build a self-improving artificial intelligence that is superior to me in every way. This AI will solve all the problems that I can’t. So it became clear to me that in building something smarter than myself, which will build something even smarter, we can eventually colonize the universe, making it intelligent.”
In order to pursue his vision of a smart all-purpose computer, Schmidhuber set up a company called Nnaisense in Lugano in 2014. The business aims to develop “large-scale neural network solutions for superhuman perception and intelligent automation” with the ultimate goal of marketing general-purpose artificial intelligences.
For Schmidhuber, it’s merely a question of time before artificial intelligence surpasses human capabilities: “The progress is inexorable, not least because so many people are fascinated by AI. We are currently living through the decades when it will become clear how intelligence works. In the not too distant future, someone will see how the already available puzzle pieces fit together and find a simple and elegant way to give the essence of practical intelligence a concrete form.”
Intelligence for the universe.