Jonathan Stray is a research fellow at the Partnership on AI, focussing on the societal side effects of optimization. He previously taught the dual Masters degree in Computer Science and Journalism at Columbia Journalism School. He has worked for ProPublica, the New York Times, and MIT Tech Review producing data-driven investigative reporting on health care, consumer debt, campaign finance, news recommendation systems, and algorithms used in the criminal justice system. He teaches data-driven methods internationally and has lectured at Columbia, Stanford, NYU, Tsinghua and Fudan universities, and publishes research on AI and data visualization in peer-reviewed journals.

He previously built several pieces of public-interest technology including Workbench, a platform for reproducible data journalism without coding, and Overview, a natural language processing and visualization system for investigating large sets of documents. He began his career in computer graphics as a research scientist for Adobe Systems, before becoming an editor at the Associated Press.

He holds a BSc. in Computer Science and Physics and an Msc. in Computer Science from the University of Toronto, and an MA in Journalism from the University of Hong Kong.