Foundational AI
At Duke, foundational AI research advances the methods and systems that make modern artificial intelligence possible, from machine learning and optimization to natural language processing and interpretable AI. By strengthening these foundations, Duke supports both cutting-edge technical discovery and the responsible, real-world application of AI.
Research Highlight: New AI Model Makes Drug Discovery Faster, Smarter, and More Transparent
A new algorithm could help researchers better predict how molecules bind to proteins — an essential step in designing more effective drugs to treat a wide range of diseases.
The Interpretable Machine Learning Lab
Duke’s foundational AI work encompasses leading research in interpretable and transparent machine learning through Cynthia Rudin’s Interpretable Machine Learning Lab, which has shaped how high-stakes AI systems are designed and evaluated in areas like healthcare and criminal justice.
Scholar spotlights: Meet the researchers building AI at Duke
The AI group in the Department of Computer Science advances research spanning the theoretical foundations of AI, the principles and methods of machine learning, and the development of practical applications that address real-world challenges. Learn about core research areas and scholars in each area from the Department of Computer Science, and check out some stories on their work below.
The Duke AI Group: History
Duke’s strength in foundational AI reflects decades of leadership in the field, rooted in the early work of the Duke AI Group established in 1973. This legacy continues to shape the university’s contributions to the development of rigorous, methodologically grounded AI systems.
Core Areas of Research
Duke’s foundational AI research spans a range of core areas that define how intelligent systems are built, understood, and applied.
Applies foundational AI methods to address complex societal challenges, from public health and education to sustainability and social equity.
Develops methods that enable machines to interpret and reason about visual data, including images, video and real-world environments.
Integrates perception, planning and control to create intelligent systems capable of acting and adapting in the physical world.
Explores how ethical principles and moral reasoning can be integrated into AI systems to guide decision-making and societal impact.
Develops algorithms for efficiently exploring complex problem spaces and identifying optimal or near-optimal solutions.
Studies how algorithms can model, analyze and design collective decision-making processes in domains such as voting, markets and resource allocation.
Advances algorithms and models that allow systems to learn from data, identify patterns and improve performance across tasks..
Natural Language Processing (NLP)
Focuses on enabling machines to understand, generate and interact with human language in accurate and meaningful ways.
Integrates perception, planning and control to create intelligent systems capable of acting and adapting in the physical world.
The Latest on Foundational AI at Duke
Study AI at Duke
Duke offers a variety of academic programs, specializations and opportunities for undergraduate and graduate students to study AI at Duke. Explore our academic programs page to learn more.