publications

2026

1. Constructing the Visual Media Forensics Ontology: A Systematic Review and Expert-Refined Framework

   Y. Kelly Wu, Saniat Javid Sohrawardi, Candice R. Gerstner, Ersin Uzun, and Matthew Wright

   2026

   In Submission
2. Beyond the Naked Eye: Empirical Study of How People Perceive, Detect, and Respond to AI-Manipulated Videos

   Kaniz Fatima^*, Y. Kelly Wu^*, and Ersin Uzun

   2026

   Conditionally accepted at CHI 2026

2025

1. Cross-lingual Hybrid and Abstractive Summarization Automatic with Attribution for the neuCLIR 2024 Pilot Report Generation Track

   John M Conroy, Razieh Fathi, Daria Smylova, and Y. Kelly Wu

   In NIST SP1329 The Thirty-Third Text REtrieval Conference (TREC 2024), 2025

   Website
2. Verification AI in the Newsroom: A Cross-Cultural Study of Journalists’ Use of Deepfake Detection Tools

   Saniat Javid Sohrawardi, Y. Kelly Wu, and Matthew Wright

   In PROMISE – PROMoting AI’s Safe usage for Elections, 2025

   Abs DOI

   This chapter examines how journalists in the United States and Bangladesh perceive and utilize deepfake detection tools in their news verification workflows. Through a semi-structured, scenario-based role-play study, we investigate the factors influencing journalists’ adoption of these tools, their placement within the verification process, and the potential biases associated with their use. Our findings show that while journalists recognize the potential of deepfake detection tools, their usage is not yet standardized or universally embraced. The tools are often employed midway through the workflow, only after initial assessments based on traditional methods prove inconclusive. Factors influencing tool usage include uncertainty about content authenticity, the perceived importance of the news story, and explicit speculations of deepfakes. The study also revealed instances of automation bias and confirmation bias among journalists, emphasizing the need for a balanced approach that combines technological tools with traditional journalistic methods and critical thinking skills. We conclude by discussing the implications of our findings for the development and deployment of deepfake detection tools that are effective, reliable, and ethically sound.

3. Understanding and Empowering Intelligence Analysts: User-Centered Design for Deepfake Detection Tools

   Y. Kelly Wu, Saniat Javid Sohrawardi, Candice R. Gerstner, and Matthew Wright

   In Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems, Yokohama, Japan, 2025

   Abs DOI

   Intelligence analysts must quickly and accurately examine and report on information in multiple modalities, including video, audio, and images. With the rise of Generative AI and deepfakes, analysts face unprecedented challenges, and require effective, reliable, and explainable media detection and analysis tools. This work explores analysts’ requirements for deepfake detection tools and explainability features. From a study of 30 practitioners from the United States Intelligence Community, we identified the need for a comprehensive and explainable solution that incorporates a wide variety of methods and supports the production of intelligence reports. In response, we propose a design for an analyst-centered tool, and introduce a digital media forensics ontology to support analysts’ interactions with the tool and understanding of its results. We conducted a study grounded in work-related tasks as an initial evaluation of this approach, and report on its potential to assist analysts and areas for improvement in future work.

2024

1. Dungeons & Deepfakes: Using scenario-based role-play to study journalists’ behavior towards using AI-based verification tools for video content

   Saniat Javid Sohrawardi, Y. Kelly Wu, Andrea Hickerson, and Matthew Wright

   In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems, Honolulu, HI, USA, 2024

   Abs DOI

   The evolving landscape of manipulated media, including the threat of deepfakes, has made information verification a daunting challenge for journalists. Technologists have developed tools to detect deepfakes, but these tools can sometimes yield inaccurate results, raising concerns about inadvertently disseminating manipulated content as authentic news. This study examines the impact of unreliable deepfake detection tools on information verification. We conducted role-playing exercises with 24 US journalists, immersing them in complex breaking-news scenarios where determining authenticity was challenging. Through these exercises, we explored questions regarding journalists’ investigative processes, use of a deepfake detection tool, and decisions on when and what to publish. Our findings reveal that journalists are diligent in verifying information, but sometimes rely too heavily on results from deepfake detection tools. We argue for more cautious release of such tools, accompanied by proper training for users to mitigate the risk of unintentionally propagating manipulated content as real news.