Richard S. Sutton, Amii Fellow and Canada CIFAR AI Chair, honoured with top prize in computing science alongside long-time collaborator Andrew Barto for their foundational work on reinforcement learning

Amii Chief Scientific Advisor Richard Sutton is known worldwide for how he has shaped research into artificial intelligence.

His ideas built the foundations for reinforcement learning, and his continued dedication to his works has inspired other researchers across the globe.

Now, Sutton is being honoured with the highest award in computer science for his impact on modern AI research — and its future. He, along with his long-time collaborator Andrew Barto, have been awarded the A.M Turing Award from the Association of Computing Machinery, often called the “Nobel Prize for Computer Science”.

So, it is fair to call Rich Sutton an authority in artificial intelligence.

Although, he never would.

“There are no authorities in science,” he says.

“No one needs anyone’s permission to question things.”

That rebellious approach has been a hallmark of Sutton’s remarkable career. He’s not one with a deep reverence for prestige or unquestionable authority. For Sutton, ideas matter above all. And good ideas can come from anywhere.

And one idea has been the driving force behind Sutton’s work for decades: understanding the mind.

A Lifelong Quest to Understand Intelligence

At times, Sutton has issues with the term “artificial intelligence.” Instead, he sees himself as simply studying intelligence itself — developing insights into the workings of the mind, whether that mind is built out of silicon or grey matter.

Building artificial intelligence is just one path, for him, to better understand all types of minds.

He argues that understanding intelligence is "one of the few great goods in the universe.” It would lead to a better understanding of ourselves as humans and our place in the world. And it would be a necessary step forward to solving many of the most pressing problems facing society, leading to massive benefits for humans as a whole.

Sutton started his quest to decipher intelligence in the 1970s, as an undergraduate at Stanford. At the time, computers were still a new technology, and paths to study artificial intelligence were rare. Sutton started studying neural psychology but found himself dissatisfied with the discipline. Sutton places a lot of importance on clear, fresh thinking — he encourages bold thinking and challenges the status quo. So he was frustrated with what he felt was calcified thinking within psychology and a resistance to new perspectives, even when they were well-supported.

After finishing his undergrad, Sutton moved on to the University of Massachusetts at Amherst, where he received first a Masters, and then a Doctorate of Philosophy in Computer Science.

Reinforcement Learning: An Introduction

It's a common cliche to say that someone "wrote the book" on a subject to express their expertise.

But for Sutton, it's actually true.

The power of learning by trial and error was a thread that stretched back to the earliest days of artificial intelligence research — and even further back, with its roots in animal behaviour studies.

Sutton wove that thread together with approaches to optimal control, culminating in his partnership with Andrew G. Barto, where they published Reinforcement Learning: An Introduction in 1998 - the book that would lay out the foundations of reinforcement learning, planting the seed for a powerful new approach to machine learning.

“I think it is not wrong to think Andy and I have been a fixed point in RL,” he says.

“We got together and said ‘This should be a field.’ Did it for a little while, and then decided ‘hey, let’s write a book and make it a field.’

And that’s what happened. That original work has been cited nearly 80,000 times in the 27 years since it was published. And reinforcement learning has grown into a promising field of AI research with vast possible applications, especially in complex and dynamic environments. Reinforcement learning has fueled great advancements in robotics, large language models, industrial control, recommendation systems and a host of other areas.

Amii, where Sutton is Chief Scientific Advisor and Fellow, has become the worldwide leader in reinforcement learning. It supports fundamental research in RL and other AI fields and works to apply that work in real-world applications to benefit everyone.

While the current impact of reinforcement learning is impressive, Sutton thinks it is only scratching the surface. He thinks it is the key to that final grand scientific prize: finally understanding our own minds.

In 2022, Sutton and Amii Fellows & Canada CIFAR AI Chairs Michael Bowling and Patrick Pilarski presented an ambitious pathway to develop Artificial General Intelligence (AGI) — a machine intelligence that can operate at near human levels, which Sutton calls "the holy grail" of AI research. Reinforcement learning, with its ability to teach agents to react and adapt to an unpredictable, dynamic world, is at the very core of that approach.

And in 2023, Sutton partnered with Keen Technologies co-founder and celebrated software engineer John Carmack with the goal of accelerating research into Artificial General Intelligence.

“Get a notebook, write things down.”

Over the course of his remarkable career, Sutton has revealed many sides of himself. Reinforcement learning visionary. Public speaker. Community builder. Cancer survivor. Nature lover. Chess player. Dance floor phenomenon.

But perhaps his most lasting impact will come from his role as a mentor.

Sutton has supervised over a dozen PhD students, many of whom continue to advance the future of AI. David Silver, the lead researcher behind AlphaGo, studied under Sutton. So did Doina Precup, who leads DeepMind’s Montreal office, Adam White, an Amii Fellow and startup founder who applies RL to improve clean water systems, and Amii CEO Cam Linke.

Those who have worked with Sutton know it’s common to see him tucked away in alcoves and waiting areas, hunched over in deep conversation with a student, drinking a Diet Coke while gently prodding and challenging their ideas and reforging them into a stronger form.

It's in these interactions that Sutton's obsession with fostering and developing new ways of thinking is most apparent. He began a summer tradition at Amii that he calls 'Tea Time Talks,' a series of short lectures that give students and other researchers a platform to share and test developing ideas — always with Rich in the audience, loaded for the Q&A.

He is forever challenging his students to explore new ideas. He encourages everyone who studies under him to keep a notebook.

“If you want others to care about what you think, then start by caring yourself. Get a notebook, write your thoughts down, challenge them, and develop them into something worth sharing.”

His own notebooks are filled with reflections on intelligence, learning, and other research ideas. He teaches his students that great ideas don’t arrive fully formed—they grow through curiosity, iteration, and engagement. Just as reinforcement learning systems improve through experience, our understanding of intelligence deepens when we interact with our thoughts, refine them, and push them forward.

"I still want to do amazing things."

Sitting in the cafe on Amii’s second floor on the eve of the A.M. Turing Award announcement, Sutton reflects on the connection between research past and present. He notes that Alan Turing was likely the first person to discuss artificial intelligence publicly, telling a London audience in 1947 that “What we want is a machine that can learn from experience.”

That idea has guided Sutton’s remarkable career for the past half-century. Artificial intelligence as a concept is much different than it was in Turing’s day: no longer theoretical, it is already providing real benefits to people’s lives.

But as much progress has been made, Sutton thinks it is nothing compared to what is to come.

“It’s a marathon, not a sprint. We still have a long way to go. I think the really impactful aspects of AI are still yet to come.”

Some might see an honour like the A.M. Turing Award as the culmination of a celebrated career. But Sutton doesn’t see it as a finish line. More of a mile marker on his way to his ultimate goal.

“I want to live up to it,” he says. ”I really do want to figure out how the mind works in a deep sense. Deeper than what we've done so far.

So I'm 67 years old, but I want to still try to do some amazing things. Which is an honorable desire, and I don't see why I can't succeed.”

And, Above All, Ideas Matter

At its core, Sutton’s quest to understand intelligence isn’t just about building AI. It’s about exploring how ideas take shape, adapt, and lead to discovery.

Just as reinforcement learning systems learn by interacting with the world, Sutton reminds us that progress comes from engaging with ideas, refining them, and allowing them to develop into something that moves our intelligence forward.

And that, more than anything, is his legacy.