How AI is emerging as a game changer in academia and sports

As global interest in sporting events continues to surge, the sports market reached an impressive $512.14 billion in 2023. AI is poised to further drive this growth by enhancing sports engagement, analysis, and forecasting, while delivering real-time data to players, broadcasters, sponsors, and fans. At the same time, academia is grappling with the challenges and opportunities of integrating AI into education, particularly with tools like ChatGPT, which can solve assignments with minimal input. These intersections of AI with sports and academia were the focus of a recent fireside chat on August 23, 2024, featuring Vishal Misra, Professor and Vice Dean of Computing and Artificial Intelligence at Columbia Engineering, and Harsh Jain, CEO of Dream Sports and a Columbia Business School alumnus. The event was organized in collaboration with Dream Sports and the Tech Entrepreneurs Association of Mumbai.

AI's Role in Cricket and Beyond

The speakers discussed AI's gradual yet significant integration into the sports industry, particularly in cricket. Misra, co-founder of Cricinfo (now owned by ESPN),  wrote technical papers in 2018 that applied state-of-the-art machine learning methods to model cricket matches. Since then, he has joined the management of the San Francisco Unicorns, where he works with a team of other cricket-obsessed scientists, including co-owner Anand Rajaraman, an IIT Madras Gold Medalist and PhD from Stanford. At the Unicorns, Misra has been implementing his advanced techniques along with other cutting-edge approaches in modern professional cricket. His work in applying AI to cricket has been pioneering, resulting in the development of a highly accurate predictive tool. The effectiveness of this tool was notably showcased during the 2024 ICC World Men's T20 World Cup, where it accurately predicted the outcome of a high-profile India-Pakistan match. The success and potential of these innovative methods have garnered attention from prominent cricket organizations, including Cricket Australia.

Optimizing Business Operations

Jain detailed how AI has enabled Dream Sports to handle vast amounts of data and personalize user experiences at scale, significantly reducing financial risks. He explained that the company uses machine learning and reinforcement learning to predict user participation in contests, improving accuracy from a 10% error margin to just 0.1%. This precision is crucial to the company's business model, which relies on collecting service fees from contest participants.

Industry-Academia Collaboration

The discussion underscored the importance of collaboration between academia and industry. Jain highlighted the newly established Columbia-Dream Sports AI Innovation Center, designed to drive innovation in sports technology. He emphasized the mutual benefits of such partnerships, with academia gaining access to large-scale data and real-world problems, while industry can quickly validate and implement research findings.

Integrating AI in Research and Education

Misra, recently appointed as Vice Dean of Computing and Artificial Intelligence at Columbia Engineering, outlined the university's efforts to integrate AI across research, teaching, and internal processes. He discussed the potential of AI tools like ChatGPT to enhance learning while cautioning against over-reliance. Additionally, Columbia is leveraging AI to streamline business processes and expand its computational infrastructure to support advanced AI research.

Innovating with Natural Language Interfaces

Misra also reflected on his work with natural language interfaces, developed to simplify access to complex data. His "Ask Here First" technology, based on ChatGPT-3, has already been implemented in ESPNcricinfo's "AskCricinfo" search engine. He is now working to commercialize this technology for broader applications, such as e-commerce and finance, allowing users to interact with systems using natural language instead of code.

Regulating AI: Models vs. Applications

The conversation also addressed the need for AI regulation, with Misra arguing that the focus should be on applications rather than the models themselves. He cautioned that over-regulation could stifle innovation, while under-regulation might fail to address the risks associated with AI's rapid development.

The Future of AI and Human Creativity

The discussion concluded with reflections on AI's potential to free humans from routine tasks, enabling greater creativity and strategic thinking. Both speakers emphasized that AI should be viewed as a tool for solving genuine problems rather than a trend to chase. Jain encouraged aspiring entrepreneurs to focus on areas where AI can make a meaningful difference, rather than simply jumping on the bandwagon.