MIT Schwarzman College of Computing Dean Dan Huttenlocher says bringing disciplines together is the best way to meet the challenges and opportunities provided by rapid advances in computing.
What does it mean to be human in a world where artificial intelligence agents make decisions that influence human behavior? That’s a big question with no easy answers, and it’s been bothering Dan Huttenlocher SM ’84, PhD ’88, dean of the MIT Schwarzman College of Computing, for a while.
“AI improvements will occur, but the destination to which we will arrive is up to us, and it is far from certain,” says Huttenlocher, who is also the Henry Ellis Warren Professor in the Department of Electrical Engineering and Computer Science.
Huttenlocher, together with former Google CEO Eric Schmidt and elder statesman Henry Kissinger, recently published “The Age of AI: And Our Human Future,” a book that delves into some of the issues raised by the emergence of AI. Huttenlocher and his co-authors say that to get there, we need much more informed dialogue and much more multilateral dialogue. “Our hope is that the book will pique the interest of people from a variety of backgrounds,” he says.
Huttenlocher, who has been the college dean for about two and a half years, doesn’t just speak the talk when it comes to interdisciplinarity. He is guiding the college as it integrates computer science into all fields of study at MIT while also educating students on how to utilize powerful tools like artificial intelligence ethically and responsibly.
This purpose is being carried out in part by two campus-wide initiatives that Huttenlocher is particularly pleased about: the Common Ground for Computing Education and the Social and Ethical Responsibilities of Computing (SERC). SERC is supported by a number of research and scholarly initiatives, including AI for Health Care Equity and the Research Initiative to Combat Systemic Racism. The Common Ground initiative encourages the development of cross-disciplinary courses that integrate computing into other fields of study, whereas the SERC initiative provides tools to help researchers, educators, and students understand how to conceptualize issues related to computing’s impacts early in the research process.
“When I was a graduate student, you worked on computer vision with the expectation that it would be a research problem for the rest of your life,” he explains. “Research challenges now have practical applications in computing-related disciplines virtually instantly.” The social consequences and ethical implications of computing must be considered from the start, not after the fact. “
Developing an interest in a new field
Huttenlocher, a deep thinker from a young age, began considering topics at the junction of human intellect and computing when he was a teenager.
With a knack for arithmetic, the Chicago native learned to program before entering high school, which was unusual in the 1970s. His parents, both academicians who investigated facets of the human mind, had an impact on his future course. His father was a neurologist at the University of Chicago Medical School who investigated brain development, and his mother was a cognitive psychology professor at the same university.
In an attempt to combine the two fields, Huttenlocher majored in computer science and cognitive psychology as an undergraduate at the University of Michigan. When it came time to apply to graduate school, he discovered the ideal fit for his two interests in the burgeoning subject of artificial intelligence and enrolled at MIT.
He studied speech recognition, object recognition, and computer vision while getting his master’s and PhD in 1984 and 1988, respectively. He became captivated by the ability of machines to immediately comprehend their surroundings. Huttenlocher was also lured in by the brisk entrepreneurial activity that was exploding around Cambridge at the time. He spent his summers interning at Silicon Valley startups and small Boston-area tech firms, which piqued his interest in the field.
“I grew up in an academic family with a healthy skepticism about following in my parents’ footsteps.” So, when I graduated, I wasn’t sure if I wanted to pursue an academic career or not. And, to be honest, I’ve felt conflicted about it ever since. For better or worse, I’ve frequently found myself doing both at the same time, he says.
Significant issues with a direct impact
Huttenlocher joined Cornell University’s computer science faculty in 1988, and he also worked at the Xerox Palo Alto Research Center (PARC), where he had interned as a graduate student. When Cornell was in session, he taught computer science courses and worked on academic research projects, and he spent his summers at Xerox PARC, as well as one day per week consulting remotely. (Long before Zoom, he notes, distant communication was “still pretty problematic.”)
“I’ve long wanted to connect the deeper, broader problems that we tend to try to solve in academics with a more direct and immediate impact on people,” he adds. “Spending time at Xerox PARC and Cornell was a fantastic way to achieve that.”
“I grew up in an academic household and had a healthy skepticism of following in my parents’ footsteps. So when I graduated, I wasn’t quite sure if I wanted an academic path or not. And to be honest, I’ve been a little bit ambivalent about it ever since. For better or worse, I’ve often ended up doing both at the same time,”
Huttenlocher
Huttenlocher began his research career by pursuing a more algorithmic approach to solving computer vision problems, rather than the more prevalent generic optimization methodologies of the time. Some of the approaches he and his colleagues invented, such as employing a graph-based representation of a picture, are still in use today, more than 20 years later.
Later, he and his colleagues conducted some of the first research on how communities interact on social networks. They analyzed LiveJournal, a major social networking site in the early 2000s, prior to Facebook. Their research found that a person’s proclivity to join an online group is influenced not just by the number of friends they have in that community, but also by how those friends are related to one another.
In addition to research, Huttenlocher was passionate about bridging gaps between disciplines. In 2009, he was appointed dean of Cornell’s interdisciplinary Faculty of Computing and Information Science. Three years later, he brought his bridge-building skills to New York City as the founding dean of Cornell Tech, a new graduate school on Roosevelt Island.
He describes his work as a significant challenge as well as an exceptional opportunity to construct a campus that links education in computing-related areas with New York’s booming tech community.
In some ways, the post prepared him well for his role as founding dean of the MIT Schwarzman College of Computing, the Institute’s most important institutional reform since the early 1950s.
“I believe this location is one of a kind. MIT has its own distinct culture. It is a distinct location in the positive sense of the word. When it comes to problem solving, people here are crazy interested and very collaborative. “It’s incredible to have the opportunity to help develop something new at MIT, something that will be essential not only for the institute, but also for the country and the globe,” he says.
Making connections
While Huttenlocher was managing the establishment of Cornell Tech, he was simultaneously building ties throughout New York City. Before the Roosevelt Island facility was established, the school rented space at Google’s Eighth Avenue headquarters, where he met then-Google CEO Eric Schmidt. The two liked discussing (and even disagreeing about) the benefits and drawbacks of artificial intelligence. At the same time, Schmidt was talking about AI with Henry Kissinger, whom he had met at a conference. By chance, the three met and began discussing AI, which led to an article in The Atlantic and, eventually, the book.
When we started talking about these problems, we realized that the broader historical and philosophical background for an AI age hasn’t been looked at very much. When people look at social and ethical issues in computers, they are usually focused on the current problem, which is essential, but we believe this larger framework is equally crucial, “he says.
Huttenlocher also feels a sense of urgency when it comes to AI problems.
Advances are occurring at such a quick pace that educational institutions are under enormous pressure to catch up. He believes that computing should be incorporated into academic courses as part of their intellectual fabric, especially as AI becomes more prevalent in everyday life. This emphasizes the importance of the college’s work as well as the challenges it will face in the future.
Huttenlocher, who has found himself in the middle of a veritable Venn diagram of disciplines since his college days, has wholeheartedly welcomed the challenge.
“These difficulties should not be addressed solely by computer scientists or engineers.” However, they should not be examined solely by social scientists or humanists, he adds. “We definitely need to bring together different communities.”
