Body-Worn Camera Research
Stanford researchers have developed methods to analyze body-worn camera footage using AI tools including natural language processing, machine learning, and automatic speech recognition.
Embedded Files
The First 45 Words
The First 45 Words
Jennifer Eberhardt on her team's research analyzing body-worn camera footage from nearly 600 traffic stops, which found that the first words an officer speaks can predict with striking accuracy whether the encounter will escalate to a handcuffing, search, or arrest.
Featured Study
Featured Study
PNAS Nexus, Volume 3, Issue 9, September 2024
Can training officers on procedural justice actually change how they speak to the public? Using body-worn camera footage as data, the researchers found that after training, Oakland officers were more likely to explain the reason for a stop, express concern for drivers' safety, and offer reassurance—demonstrating that these methods can both measure and drive change in police-community interactions.
Nicholas P. Camp, Rob Voigt, MarYam G. Hamedani, Dan Jurafsky, Jennifer L. Eberhardt
Research by Topic
Research by Topic
Racial Disparities in Officer Communication
Racial Disparities in Officer Communication
This line of research revealed that officers communicate differently with community members of different races—not only in their word choices but also in their tone of voice and in the earliest moments of an encounter, before the driver has had the chance to say much at all.
Rho, Eugenia H., Maggie Harrington, Yuyang Zhong, Reid Pryzant, Nicholas P. Camp, Dan Jurafsky, and Jennifer L. Eberhardt. “Escalated Police Stops of Black Men Are Linguistically and Psychologically Distinct in Their Earliest Moments.” Proceedings of the National Academy of Sciences 120, no. 23 (2023): e2216162120. https://doi.org/10.1073/pnas.2216162120
Camp, Nicholas P., Rob Voigt, Dan Jurafsky, and Jennifer L. Eberhardt. “The Thin Blue Waveform: Racial Disparities in Officer Prosody Undermine Institutional Trust in the Police.” Journal of Personality and Social Psychology 121, no. 6 (2021): 1157–1171. https://doi.org/10.1037/pspa0000270
Voigt, Rob, Nicholas P. Camp, Vinodkumar Prabhakaran, William L. Hamilton, Rebecca C. Hetey, Camilla M. Griffiths, David Jurgens, Dan Jurafsky, and Jennifer L. Eberhardt. “Language from Police Body Camera Footage Shows Racial Disparities in Officer Respect.” Proceedings of the National Academy of Sciences 114, no. 25 (2017): 6521–6526. https://doi.org/10.1073/pnas.1702413114.
Computational Tools for Analyzing Police Encounters
Computational Tools for Analyzing Police Encounters
These studies developed and refined the technical methods—from speech recognition and speaker identification to dialog act detection—that make large-scale computational analysis of body-worn camera footage possible.
Field, Anjalie, Prateek Verma, Nay San, Jennifer L. Eberhardt, Dan Jurafsky. “Developing Speech Processing Pipelines for Police Accountability.” INTERSPEECH 2023 (2023): 1229–1233. https://doi.org/10.21437/Interspeech.2023-2109.
Prabhakaran, Vinodkumar, Camilla Griffiths, Hang Su, Prateek Verma, Nelson Morgan, Jennifer L. Eberhardt, and Dan Jurafsky. “Detecting Institutional Dialog Acts in Police Traffic Stops.” Transactions of the Association for Computational Linguistics 6 (2018): 467–481. https://doi.org/10.1162/tacl_a_00031.
Summit on AI, Body-Worn Cameras, and the Future of Policing
Summit on AI, Body-Worn Cameras, and the Future of Policing
In 2024, researchers from across Stanford—including the Graduate School of Business, School of Humanities and Sciences, School of Engineering, and Stanford Law School— convened policymakers, law enforcement leaders, technologists, and policy experts to explore how advances in AI, combined with more widespread camera adoption and proven research approaches, present a renewed opportunity to harness body-worn camera footage to improve policing in America.
Andrew Broadhead
Page updated
Report abuse