In the video above, Columbia Business School Dean Costis Maglaras, a leading expert in data science, discusses how AI is revolutionizing the school’s curriculum, classroom instruction, and faculty research.
Watch the interview with Dean Maglaras above and read the transcript below.
Learn more about AI@CBS.
CBS: How did your AI journey begin?
Dean Costis Maglaras: It goes back a very long time. In fact, when I was doing my PhD in the early '90s, I spent some time in the AI Lab at Stanford in the computer science department. Back then, we were trying to teach computers to recognize images, and I can tell you, it was incredibly painful and a very difficult problem. Fast forward 30 years, and these machines and models have become exceptionally powerful. Our ability to compute has improved by a factor of a billion from the time I was in grad school to today. That means if you spend a dollar today, you can do a billion times more calculations than you could have done 30 years ago with the same dollar.
We now have incredibly powerful computers, abundant data, and sophisticated models that bring these two things together to create truly innovative technologies. My journey started very early, and in the last four or five years, we've seen AI penetrate not only academic research but also our day-to-day lives—from self-driving cars to everyday online interactions. The growth of large language models in the last couple of years is exciting, and it’s fascinating to imagine what will happen in the future.
CBS: How is CBS embracing AI?
Dean Maglaras: In the past three or four years, we’ve introduced several courses. One of them was on breakthrough technologies, which I co-taught with the Dean of the School of Engineering, Shih-Fu Chang. A lot of that course was about deep learning, AI models, robotics, self-driving cars, and more. We're using AI to teach technology strategy, and we offer courses specifically on how these models work, when they are effective, and when they aren’t. We even have courses where students gain hands-on experience with AI, such as learning how to prompt and use these models.
There is a lot of curricular innovation in how we teach about these technologies—how they work, when they are effective, and how we expect them to penetrate industries in the next two, four, or six years. We also focus on using these tools to enhance our teaching of all subjects, including thinking about how AI will change marketing, investment management, accounting, reporting, and auditing. In some ways, it's multi-pronged: we teach about the tools, we use the tools, and we imagine how the tools will disrupt different jobs and industries. We've been working diligently on this over the last three or four years, and many of our students are now getting significant exposure to these ideas. Some of these concepts have even been integrated into the Core Curriculum, meaning that every single student gets exposed to these ideas in their first semester.
CBS: What’s unique about CBS’s approach to AI?
Dean Maglaras: Thirteen years ago, we updated our Core Curriculum to include a business analytics class that covers machine learning and its real-world applications. This means we’ve been leading in this space for quite some time, particularly among peer schools. Our deep connection with the School of Engineering allows us to expose students to leading faculty who discuss cutting-edge topics like robotics and deep learning models. This gives our students a unique perspective on where these technologies may go in the next three, five, or ten years.
CBS: How is AI driving faculty research?
Dean Maglaras: AI is also becoming central to the research we do here. One area focuses on foundational research, such as improving AI models and training large language models to incorporate business objectives. We’re also exploring reinforcement learning, which involves improving systems based on actions and consequences in real-world applications. Another research stream uses AI’s power to analyze and categorize text at a larger scale.
It’s not just about the tools, though—AI’s impact on people, work, and leadership is crucial. We’re thinking about how AI will change management practices and its broader implications for society. We’re also exploring AI applications in fields like investment, climate modeling, and accounting.
CBS: How will AI reshape business and society?
Dean Maglaras: In the next decade, we’ll start seeing concrete AI applications. Some will be in our online interactions, others in areas like drug discovery, which will be disrupted and enhanced by AI models. Over the next 10 to 20 years, we’ll see advancements in robotics, including self-driving cars, and AI will become a part of our daily lives, affecting the jobs our students take after graduation and creating entirely new products and solutions.
AI will play an important role in addressing climate challenges, and it will be a driver of profound change over the coming decades. However, this change will be gradual. People and large organizations tend to resist change. For example, in 2000, we had the technology to switch to online banking, but most people continued visiting their local branches. It wasn’t until COVID-19 forced us to stop visiting branches that we fully adopted online banking—something we could have done 20 years earlier. This resistance to change will affect how quickly AI is deployed, but I believe the next 10, 20, or 30 years will be an exciting time for AI.
