Prof. Karrie Karahalios

Professor of Media Arts and Sciences

Areas of Interest and Expertise

Social Visualization
Sociable Media

Research Summary

Dr. Karrie G. Karahalios is an internationally recognized computer scientist whose research bridges computing, society, and systems design. In September 2025, she rejoined the MIT Media Lab as a Full Professor of Media Arts and Sciences, bringing a distinguished record of scholarship focused on algorithmic accountability, human-computer interaction, and responsible technology.

Dr. Karahalios holds four degrees from MIT: an SB and ME in Electrical Engineering and Computer Science, and an SM and PhD from the Media Lab, completed in 2004. Her doctoral work helped launch an interdisciplinary career that integrates technical expertise with social, legal, and ethical inquiry.

Prior to her return to MIT, she served as a Professor in the Department of Computer Science at the University of Illinois Urbana-Champaign (UIUC). She held affiliate appointments in Electrical and Computer Engineering, the Coordinated Science Laboratory, the School of Information Sciences, and the Unit for Criticism and Interpretive Theory. She also co-founded and co-directed the Center for Just Infrastructures, a cross-disciplinary initiative focused on inclusive, community-centered technology development.

Dr. Karahalios is widely recognized for co-developing the concept of algorithmic audits, a now-standard framework for diagnosing bias and discrimination in opaque digital systems. Her work has examined how algorithms influence user behavior and shape outcomes in contexts such as social media feeds, housing markets, health platforms, and civic technologies. Her research draws on computer science, sociology, law, psychology, and design to inform the development of more transparent and accountable systems.

In 2016, Dr. Karahalios was a plaintiff in a landmark ACLU lawsuit challenging aspects of the Computer Fraud and Abuse Act (CFAA), helping to establish legal protections for researchers examining algorithmic bias in public-facing systems.

Recent Work