HUMAN BRAINS ARE TOO PUNY and 15 more reasons why top researchers are obsessed with AI

I've been very interested in computers since I was a little kid. My older brother taught me computer programming when I was five years old. I was really drawn to it because it gave me control over the computer and it was really creative.

Once you learn how to program you can do whatever you want with a computer. There are a lot of creative opportunities. It's also a really addictive kind of problem solving. So I knew from a pretty young age that I wanted to work with computers.

But then as I got older I got really frustrated by how slowly humanity was solving the fundamental mysteries of the universe. I thought the bottleneck here is that our brains are just too puny. It's too hard to think about these really big problems with our little brains so what we need to do is we need to augment our brains with something that will make them smarter. We need to make computers so smart that they can help us solve these big problems.

Shimon Whiteson is an associate professor at the Informatics Institute at the University of Amsterdam.

I've always been fascinated by understanding how things work. If I had gone on to study whatever I found most intriguing in high school, it'd probably have been physics. But with the field of physics already so far along, it just seemed engineering had more potential to lead to doing something that'd have tangible results within my lifetime.

Within engineering, artificial intelligence quickly became most fascinating to me. Building a system that can do (somewhat) intelligent reasoning seemed like it'd open up a lot of possibilities, and also downright intriguing.

Pieter Abeel is a computer scientist at University of California, Berkeley.

Originally I wanted to become a professional tennis player. When I realized that's not what I was going to be, I wanted to build a robotic system, like a robotic buddy who could play tennis with me.

In order for me to build a robot like that, capable of playing tennis, I had to give it plenty of intelligence ... it has to be able to think and then move accordingly. As I started wanting to build that robot at the Berkeley undergraduate school and [at] MIT, I started to realize I had to study a lot of AI.

Yoky Matsuoka is the former vice president of technology at Nest, a Google-owned company that makes smart thermostats.

I got into AI very early, about the time when I started my masters. I was looking for a subject and that's when I read about neural networks.

Since then it's been love at first sight ... it's been an intense passion for me.

I guess I had been a reader of science fiction as an adolescent and my mind was full of dreams of intelligent machines.

Yoshua Bengio is a computer scientist at University of Montreal.

I started working on computer science as a hobby when I was at school. I guess you would say about ninth grade. I did a computer science A level and I also wrote a chess program so I convinced Imperial College to let me use their supercomputer to write a chess program.

Then I read a couple books about AI, "Problem Solving Methods in Artificial Intelligence" by Nils Nilsson and "The Thinking Computer" by Bert Raphael. It seemed to me that AI was one of the most interesting problems.

I didn't at that time think I would ever do it professionally because I was going to be a theoretical physicist. But then when it came time to choose a graduate school I applied to both computer science and physics and in the end, I chose AI.

Stuart Russell is a computer scientist at the University of California, Berkeley.

When I was growing up I liked sci-fi books, so I read a lot. I really enjoyed "I, Robot" by Isaac Asimov.

In undergrad I decided that I wanted to build robots, so I majored in mechanical engineering, because that's how you build robots. That's what I thought, anyway. At the time, you needed mechanical engineering skills.

Then later on I realized that one of the hardest challenges in robotics was ... how it can understand the world through sensors and make decisions. That led me to where I am today.

Carlos Guestrin, is the CEO and co-founder of Dato, a company that builds artificially intelligent systems to analyze data.

I was a masters student over at Carnegie Mellon University, and I was part of a team that did robot soccer. We had these little robot dogs, and they would go around and they would kick soccer balls, and they would go and do competitions with these robots. That really gave me a taste of robotics, first of all, and also a taste of having robots that worked together.

From there, I started to realize that this idea of having many robots work together could be useful for things like medicine, for things like search and rescue, for things like cleaning up an oil spill. I realized the potential of this and went into robotics.

Sabine Hauert is a roboticist at Bristol University.

This goes right back to high school. There's a book that was very popular at the time by Douglas Hofstadter called "Godel, Escher, Bach."

In the book, Hofstadter made these amazing connections between mathematics, metamathematics, art, music, and the study of computer science. When I read that, I was hooked.

"How do you build an intelligent machine?" is one of the most fundamental, intellectual questions across all the sciences.

It's a mystery like "what is the origin of the universe," or "what is the nature of space."

It's very very fundamental, and I felt like this was something that I could devote my professional life to.

Oren Etzioni is the CEO of the Allen Institute for Artificial Intelligence.

Before I was a teenager, home computers started to become available, and I used to go to the mall and play with them because I was just fascinated that they somehow knew the answers to lots and lots of different questions.

I was wondering how they do that. For my thirteenth birthday, I got a computer and kind of locked myself away for a couple of years and played with it until I felt like I understood how it worked. It's just been fascinating since then.

Michael Littman is a computer scientist at Brown University.

I've been interested in AI since I was a kid, really. I was very much drawn into science fiction, and really science fiction is what got me into the field.

Reading Asimov's robot stories, "I, Robot," I was very impressed as a young chap. I really wanted to be like Susan Calvin, who's the robot psychologist in Asimov's robot stories. She was sort of my fictional hero.

From that the obvious subject to study was computer science. So I got into computer science — studied computer science at Imperial College ... then it's sort of natural to segue from that into the kind of AI research that I'm interested in.

Murray Shanahan is a computer scientist at Imperial College.

My dad is an engineer (retired now) and got me interested in designing and building things, particularly model airplanes and electronics. I was fascinated by the mystery of intelligence as a kid, which got me interested in human evolution, epistemology and AI.

Because I like building things, I taught myself electronics and programming in high school and studied electrical engineering. Then I studied AI for my graduate studies.

Yann LeCun is the director of Facebook Artificial Intelligence Research.

I took a class when I was getting my masters degree that was in artificial intelligence. That class was really interesting to me — it was my first exposure to AI.

As part of that class, we conceptually developed a robot for a particular task — just the whole idea of how you can build intelligent systems that can solve problems in ways that are similar to how humans might solve them is, to me, just a fascinating challenge.

Lynne Parker is the division director for the Information and Intelligent Systems Division at the National Science Foundation.

I was interested in how the brain works.

My motivation for getting into artificial intelligence is to understand human intelligence and to see whether aspects of human cognition and intelligence could be duplicated in machines — back in the mid 1980s.

Then I got a PhD in the field and then I've worked it in ever since.

Bart Selman is a computer scientist at Cornell University.

I was an electrical engineering student, and I was interested in speech recognition at that time. Once I started looking at speech recognition, I basically somehow got pulled into ... actually making machines intelligent — basically showing the kinds of behaviors that humans are capable of showing.

I teach AI as a professor and I always find that it's one of those subjects that is extremely easy to get people interested in ... understanding the nature of intelligence by trying to replicate it in a computational substrate is something that everybody gets excited about. That was sort of my way of entry into the field.

Subbarao Kambhapati is a computer scientist at Arizona State University.

In graduate school, I became fascinated by the fundamental questions in [the] philosophy of science — how do we learn about the world

I wanted to know how we could create a computer program that could learn about the world. This led me to machine learning and AI.

Thomas Dietterich is the President of the Association for the Advancement of Artificial Intelligence.

Studying Experimental Psychology made me realize that the kinds of theories psychologists test were hopelessly vague — we would only understand how minds work by actually building them.

Geoffrey Hinton is a researcher at Google and computer scientist at the University of Toronto.