RESEARCH & DESIGN
2026
How AI Companies Are Handling Suicide and Self-Harm Today (Partnership on AI) - Co-authored with Claire Leibowicz: an in-depth analysis of how frontier AI companies detect and respond to users in suicide and self-harm crises, including a taxonomy of six common intervention types and open questions for the field. Builds on the Cross-Sector Primer for AI and Suicide Prevention and an expert workshop hosted by OpenAI, which I designed and facilitated through Saltern Studio.
AI and the Future of Digital Public Squares (Collective Intelligence) - Led the community moderation section of a review article on how large language models could strengthen or undermine digital public squares, developed from a convening of civil society experts and technologists.
2025
Facilitation in the AI Era: A Community Roadmap for Technologies to Support Practitioners (Jigsaw) - Lead co-author, with Ian Beacock, of a collaborative report with 13 global experts in facilitating dialogue on how AI could best support public conversations.
An Ecological Framework for AI and Community Moderation (EPIC) - Case study paper on ethnographic research and codesign with community moderators to develop LLM-based moderation tools.
Sensemaking Tools in Bowling Green, Kentucky (Jigsaw) - Led research with local leaders for a civic deliberation project with 8,000 community members; 83% reported the interactive site helped them understand viewpoints they disagreed with.
2024
Bridging Attributes for Perspective API (Jigsaw) - 0-to-1 research with online community members and moderators for classifiers that surface constructive comments, used by organizations like the NYT, Nextdoor, and Viafoura. See also the evaluation paper.
Annotator in the Loop: A Case Study of In-Depth Rater Engagement to Create a Prosocial Benchmark Dataset (AIES) - Co-author of a case study on engaging raters in depth to build a benchmark dataset for prosocial content.
Independently Testing Prosocial Interventions: Methods and Recommendations from 31 Researchers (Annals of the New York Academy of Sciences) - One of three lead co-authors of cross-disciplinary recommendations for rigorously and independently evaluating prosocial interventions online.
2021
Misinformation Interventions Are Common, Divisive, and Poorly Understood (Harvard Kennedy School Misinformation Review) - Lead author of research on public attitudes toward platform misinformation interventions.
Labels for Manipulated Media (Partnership on AI and First Draft) - Foundational user research on labeling interventions for manipulated media, published at CHI.
2019–2020
The News Provenance Project (The New York Times R&D Lab) - UX lead for a blockchain prototype for labeling photojournalism, which went on to inform the C2PA industry standard for verifying media sources.
2017–2018
Stockgrok (Bloomberg/CMU) - Led design research with people with visual impairments for an accessible data sonification tool for financial charts.
Toward Measuring Empathy in Virtual Reality (CHI PLAY) - A qualitative user research framework for measuring empathy in VR.
Can Biosignals Be Expressive? - Supported research on kinetic typography driven by biosignals.
Earlier
Pop Up Archive - Community codesign and wireframing for podcast and radio search products.