Been Kim
TL;DR Been Kim is a leading researcher in interpretable machine learning, known for creating methods that help humans understand and trust complex AI systems.
Been Kim is one of the most influential figures in the field of interpretability and explainable artificial intelligence. Her work focuses on making advanced machine learning models understandable to humans, enabling transparency, trust, and accountability. By combining rigorous research with a deep understanding of human cognition, she has helped shape how the AI community approaches interpretability.
Been Kim is a research scientist at Google Brain, where she leads groundbreaking work on creating interpretability tools for deep learning systems. She is best known for developing conceptual interpretability methods that go beyond simple visualizations, aiming instead to align explanations with how humans naturally reason about concepts.
One of her most recognized contributions is the development of TCAV, a technique that reveals how neural networks interpret high-level concepts within their internal representations. Her research explores the intersection of machine learning, cognitive science, and human-centered design, aiming to develop AI systems that collaborate effectively with people.
She is also a strong advocate for building interpretability standards, responsible AI practices, and interdisciplinary research that brings together experts from machine learning, psychology, and philosophy.
Research scientist at Google Brain, leading work in interpretable machine learning
Creator of TCAV, a widely adopted method for concept-based interpretability
Pioneering research on aligning AI explanations with human reasoning
Major contributor to frameworks for understanding and evaluating neural network decisions
Advocate for interdisciplinary research, integrating cognitive science and ML
Influential voice in responsible, transparent, and human-centered AI
