In a significant recognition for the global Indian scientific community, three researchers of Indian origin have been named among the 28 global scholars selected for the prestigious Schmidt Sciences AI2050 Fellowship for 2025. The fellowship, which provides over $18 million in funding, is dedicated to supporting groundbreaking research aimed at ensuring that artificial intelligence develops in ways that benefit humanity by the year 2050.
Meet the Indian-Origin Fellows Shaping AI's Future
The programme specifically identifies leading thinkers whose work addresses critical questions of AI safety, transparency, reliability, and long-term societal impact. The selection of Surya Ganguli, Surbhi Goel, and Krishna Pillutla underscores the vital contributions of Indian-origin researchers in guiding the global trajectory of artificial intelligence.
Surya Ganguli: Building the Scientific Foundations for Trustworthy AI
Surya Ganguli is a Professor of Applied Physics at Stanford University and also serves as the Associate Director at Stanford’s Human-Centered AI Institute. His academic credentials are impressive, holding undergraduate degrees in physics, mathematics, and electrical engineering and computer science from MIT. He completed his PhD in string theory at the University of California, Berkeley, followed by postdoctoral research in theoretical neuroscience at UCSF.
His interdisciplinary work seamlessly spans neuroscience, physics, and machine learning, with a central focus on understanding how both biological and artificial neural networks perform complex learning and reasoning tasks. His research group is renowned for developing analytic frameworks to explain phenomena like creativity, reasoning, and generalisation in large language models and generative AI systems.
Ganguli's contributions have been recognized with several major honours, including the Sloan Fellowship, Simons Investigator Award, NSF CAREER Award, and the Schmidt Science Polymath Award. Through the AI2050 Fellowship, he aims to advance the scientific foundations of explainable and reliable AI, with a particular emphasis on improving how large models refine themselves through continuous interaction and feedback.
Surbhi Goel: Designing Mathematically Grounded and Safe AI
Surbhi Goel holds the position of Magerman Term Assistant Professor in the Department of Computer and Information Science at the University of Pennsylvania. Her research directly confronts a major challenge in modern AI: the lack of formal guarantees in the behaviour of complex AI systems.
Goel specializes in the mathematical foundations of machine learning, developing rigorous theoretical frameworks to understand precisely when AI systems work reliably and, just as importantly, why they sometimes fail. Her AI2050 project is focused on a crucial goal: creating AI systems that are predictable and trustworthy. This is especially vital in areas like conversational AI, where safety and reliability are non-negotiable for user trust and ethical deployment.
Her pioneering work is instrumental in reducing the uncertainty that often surrounds AI decision-making processes, paving the way for the creation of systems that can be deployed responsibly in high-stakes real-world settings.
Krishna Pillutla: Advancing Privacy-Preserving and Responsible AI
Krishna Pillutla is an Assistant Professor and Narayanan Family Foundation Fellow at the Wadhwani School of Data Science and AI at IIT Madras. He also serves as a Principal Investigator at the Centre for Responsible AI (CeRAI). His research is centered on developing privacy-preserving, robust, and fair machine learning systems for critical applications in healthcare, finance, and public welfare.
Pillutla's expertise lies in areas like differential privacy, federated learning, ensuring robustness in generative AI, and developing methods to prevent AI models from accidentally leaking sensitive data. Before joining IIT Madras, he honed his skills as a postdoctoral fellow at Google Research, where he contributed significantly to advances in decentralised and privacy-respecting AI training methodologies.
The AI2050 Fellowship recognises his pivotal role in shaping a future of responsible AI development that consistently prioritises social benefit and strong ethical safeguards over purely technological advancement.
A Global Initiative for a Human-Centric AI Future
Since its launch in 2022, the AI2050 programme has supported 99 fellows across eight countries and more than 42 research institutions worldwide. The initiative is fundamentally designed to promote collaboration among top scientists who are working to ensure that AI systems remain aligned with human values and the public interest.
The selection of Ganguli, Goel, and Pillutla is not just an individual achievement but a testament to the growing influence and excellence of Indian-origin researchers in one of the most important technological fields of our time. Their work is set to play a crucial part in steering the global future of artificial intelligence toward greater transparency, safety, and tangible societal good.
