About

Diana Freed is an Assistant Professor of Computer Science and Data Science at Brown University. Her research areas include Human-Computer Interaction, Security, Data Science, and Artificial Intelligence. She is involved in teaching courses such as Accessible and Inclusive Cybersecurity and Privacy in Fall 2026, and Intro to Sociotechnical Systems and HCI in Spring 2027. Her work focuses on the intersection of technology and society, emphasizing security, privacy, and inclusive design.

Research topics

• Computer Science
• Political Science
• Computer Security
• Internet privacy
• Psychology
• Medicine
• Nursing
• Public relations
• Business
• Law
• Medical emergency
• Engineering
• Criminology

Selected publications

• Developmentally Safe Generative AI Environment for Youth

  2026-04-13

  articleOpen access

  Generative AI (GenAI) systems such as ChatGPT, Gemini, Character.AI, and LLaMA are becoming tightly woven into youths’ everyday lives. Young people now turn to these tools not only for homework help and creative projects, but also for companionship, identity exploration, and emotionally charged conversations. These always-available, highly responsive systems can support learning and self-expression, yet they also introduce new and poorly understood risks for cognitive, emotional, social, moral, and identity development. At the same time, caregivers, educators, designers, and policymakers often have limited visibility into how youth actually use GenAI and few shared frameworks for supporting developmentally appropriate, safe, and empowering engagement. This workshop aims to critically explore the holistic impact of GenAI on youth and to build a shared, developmentally grounded agenda for research and design. Our goals are to: (1) identify promising research approaches and ethical, participatory methods for studying youth–GenAI interactions in situ; (2) surface open challenges, hidden risks, and unintended impacts of GenAI use across diverse cultures and contexts; (3) explore community-based and longitudinal research models that connect youth, caregivers, educators, designers, and policymakers; (4) co-develop shared infrastructures and evaluation metrics to guide design, education, and policy, such as datasets, toolkits, forums, and indicators of developmental safety and well-being. We will bring together researchers and practitioners from HCI, education, child development, security and privacy, and policy to collaboratively sketch actionable strategies for developmentally safe GenAI environments that balance risk mitigation with youth agency, autonomy, and space for exploration and growth.

  PublisherDOI

• Navigating Privacy and Trust: AI Assistants as Social Support for Older Adults

  ArXiv.org · 2025-05-05 · 1 citations

  preprintOpen accessSenior author

  AI assistants are increasingly integrated into older adults' daily lives, offering new opportunities for social support and accessibility while raising important questions about privacy, autonomy, and trust. As these systems become embedded in caregiving and social networks, older adults must navigate trade-offs between usability, data privacy, and personal agency across different interaction contexts. Although prior work has explored AI assistants' potential benefits, further research is needed to understand how perceived usefulness and risk shape adoption and engagement. This paper examines these dynamics and advocates for participatory design approaches that position older adults as active decision makers in shaping AI assistant functionality. By advancing a framework for privacy-aware, user-centered AI design, this work contributes to ongoing discussions on developing ethical and transparent AI systems that enhance well-being without compromising user control.

  PublisherOA PDFDOI

• Sociotechnical Challenges in Implementing Domestic Violence Screening via Telehealth and In-Person Care: Qualitative Study on Clinicians’ Perspectives

  Journal of Medical Internet Research · 2025-07-18

  articleOpen access1st authorCorresponding

  BACKGROUND: New York State Public Health Law Section 2805-z (NYSPHL§2805-z) mandates that hospitals implement certain protocols for the identification of domestic violence (DV) to strengthen support for DV survivors. However, there are challenges within our current health care delivery settings that raise critical questions about the effectiveness and adaptability of mandated policies for DV screening. The COVID-19 pandemic significantly accelerated the adoption of telehealth services, leading to the widespread use of combined telehealth and in-person clinical care. This shift has brought challenges associated with implementing the mandated DV screening protocols. OBJECTIVE: This study aimed to explore the challenges health care providers face in implementing DV screening that complies with NYSPHL§2805-z to support effective screening practices. Considering both the telehealth and in-person environment, we aimed to identify the challenges that health care providers face in meeting policy requirements, with a focus on the technical, environmental, and social factors impeding effective DV screening. We also explored potential sources of support to address these challenges. This study sought to provide actionable insights for enhancing policy implementation and improving care for DV survivors. METHODS: We conducted interviews with health care professionals-nurses, social workers, and physicians-across New York City involved in DV screening. These interviews were analyzed using an adapted version of the Centers for Disease Control and Prevention sociotechnical model to understand how policy, organizational structures, and individual practices intersect and impact the effectiveness of DV screening, especially in telehealth settings. RESULTS: Our findings revealed gaps in awareness of the policy and inconsistent DV screening processes. Through our interviews, we identified challenges in effective DV screening and strategies to improve the screening process. We found that the shift to telehealth introduced additional layers of complexity, with challenges in ensuring patient privacy and safety. Our findings revealed a lack of clear guidelines tailored to remote DV screenings and a need for training programs to prepare health care providers for the nuances of telehealth-based DV screening, both of which are crucial for complying with policy mandates. On the basis of this analysis, we developed a list of recommendations to support health care providers in effective screening for DV. Through user feedback, we confirmed that this list is consistent with the application of NYSPHL§2805-z among health care providers. CONCLUSIONS: As telehealth care is increasingly becoming an integral component of health care, there is an urgent need to refine and reinforce DV screening policies and procedures designed to comply with NYSPHL§2805-z. This study highlights the importance of developing practical and consistent telehealth protocols, enhancing provider training, creating supportive workflows, allotting adequate resources, and fostering collaboration among stakeholders to ensure that DV survivors receive the care they need in both remote and in-person clinical settings.

  PublisherDOI

• Insights from Auditing AI Chatbots for Survivors of Domestic Violence

  Open MIND · 2025-01-01

  articleOpen accessSenior author
  PublisherOA PDFDOI

• Medical Data for Sale: Accessing Reproductive Health Information via the Data Brokerage Landscape

  IEEE Security & Privacy · 2025-07-01

  article1st authorCorresponding

  This study investigates the ease with which third parties can access and exploit intimate health data, including mental health- and abortion-related information after the Dobbs decision. Our findings demonstrate incongruities surrounding the perceived protections of HIPAA and the ready availability for the purchase of medical data.

  PublisherDOI

• From Care to Technology-Facilitated Abuse: Law and Policy Approaches to Safeguard Against AgeTech Misuse

  SSRN Electronic Journal · 2025-01-01

  preprintOpen access1st authorCorresponding
  PublisherDOI

• PROTECT: A Framework to Foster Digital Resilience for Youth Navigating Technology-Facilitated Abuse

  Social Sciences · 2025-06-16 · 7 citations

  articleOpen access1st authorCorresponding

  Youth are increasingly exposed to a broad range of technology-facilitated abuse that challenges their safety and well-being. Building on previous work that examined youth help-seeking behaviors, coping strategies, threats they encounter, and the social support systems around them, we articulate a framework— called PROTECT—Problem recognition, Reaching out, Organizing support, Training, Engaging experts, Continuous support, and Tackling safety measures—which integrates existing models of support, help-seeking, and digital skills to offer a high-level, structured approach to adults who serve as a support system to youth navigate technology-facilitated abuse. The framework unpacks social and contextual dynamics that influence help-seeking behaviors, providing a foundation for educators, advocates, health professionals, developers and other adult stakeholders to design and develop trauma-informed, timely interventions to promote resilience.

  PublisherOA PDFDOI

• Help-seeking and Coping Strategies for Technology-facilitated Abuse Experienced by Youth

  Proceedings of the ACM on Human-Computer Interaction · 2025-05-02 · 3 citations

  article1st authorCorresponding

  Technology provides youth (ages 10--17) with near-constant opportunities for learning, communication, and self-expression. It can also expose them to technology-facilitated abuse: harassment, coercion, fraud, and more. The ability of youth to navigate such abuse is crucial for their well-being and development. A recent advisory by the U.S. Surgeon General called for better support of youth, including that youth should ''reach out for help.'' However, little is known about how youth seek help or otherwise cope with technology-facilitated abuse. Through a qualitative study in the U.S., we examine how youth engage in self-reliance, seek help from others, and how others seek help on a youth's behalf. We discuss these strategies and outline opportunities for how the HCI community can better support youth who experience technology-facilitated abuse.

  PublisherDOI

• Sociotechnical Challenges in Implementing Domestic Violence Screening via Telehealth and In-Person Care: Qualitative Study on Clinicians’ Perspectives (Preprint)

  2025-03-31

  articleOpen access1st authorCorresponding

  <sec> <title>BACKGROUND</title> New York State Public Health Law Section 2805-z (NYSPHL§2805-z) mandates that hospitals implement certain protocols for the identification of domestic violence (DV) to strengthen support for DV survivors. However, there are challenges within our current health care delivery settings that raise critical questions about the effectiveness and adaptability of mandated policies for DV screening. The COVID-19 pandemic significantly accelerated the adoption of telehealth services, leading to the widespread use of combined telehealth and in-person clinical care. This shift has brought challenges associated with implementing the mandated DV screening protocols. </sec> <sec> <title>OBJECTIVE</title> This study aimed to explore the challenges health care providers face in implementing DV screening that complies with NYSPHL§2805-z to support effective screening practices. Considering both the telehealth and in-person environment, we aimed to identify the challenges that health care providers face in meeting policy requirements, with a focus on the technical, environmental, and social factors impeding effective DV screening. We also explored potential sources of support to address these challenges. This study sought to provide actionable insights for enhancing policy implementation and improving care for DV survivors. </sec> <sec> <title>METHODS</title> We conducted interviews with health care professionals—nurses, social workers, and physicians—across New York City involved in DV screening. These interviews were analyzed using an adapted version of the Centers for Disease Control and Prevention sociotechnical model to understand how policy, organizational structures, and individual practices intersect and impact the effectiveness of DV screening, especially in telehealth settings. </sec> <sec> <title>RESULTS</title> Our findings revealed gaps in awareness of the policy and inconsistent DV screening processes. Through our interviews, we identified challenges in effective DV screening and strategies to improve the screening process. We found that the shift to telehealth introduced additional layers of complexity, with challenges in ensuring patient privacy and safety. Our findings revealed a lack of clear guidelines tailored to remote DV screenings and a need for training programs to prepare health care providers for the nuances of telehealth-based DV screening, both of which are crucial for complying with policy mandates. On the basis of this analysis, we developed a list of recommendations to support health care providers in effective screening for DV. Through user feedback, we confirmed that this list is consistent with the application of NYSPHL§2805-z among health care providers. </sec> <sec> <title>CONCLUSIONS</title> As telehealth care is increasingly becoming an integral component of health care, there is an urgent need to refine and reinforce DV screening policies and procedures designed to comply with NYSPHL§2805-z. This study highlights the importance of developing practical and consistent telehealth protocols, enhancing provider training, creating supportive workflows, allotting adequate resources, and fostering collaboration among stakeholders to ensure that DV survivors receive the care they need in both remote and in-person clinical settings. </sec>

  PublisherDOI

• (Un)designing AI for Mental and Spiritual Wellbeing

  2024-11-11 · 5 citations

  articleOpen access

  With rapid advances in Artificial Intelligence (AI) impacting human health and wellbeing, scholarly discourse should focus equally on the prospective opportunities and harms of Human-AI Interaction (HAI) in Computer-Supported Collaborative Work and Social Computing (CSCW). This panel invites critical interdisciplinary discussion around the (un)designing of AI by asking: how, when, where, and why should AI (not) be involved in sociotechnical systems for mental and spiritual health and wellbeing? Motivated by functional, technical, and ethical concerns, the panel aims to ensure that: (1) progress in HAI for mental and spiritual health is informed by expertise from the respective clinical disciplines; (2) ethical and responsible design principles lie at the core of research motivations and methodologies; and (3) AI hype can be tempered by caution given its impacts on marginalized and stigmatized groups. A panel of respected experts in mental health, spiritual care, and AI will discuss CSCW topics regarding HAI in contexts of clinical practice (e.g., electronic health records, patient portals, decision-making and referral systems, technology-supported interactions during bedside care or clinical appointments) as well as social contexts beyond the clinic (e.g., social apps, online health communities and social media, and computer-mediated communication in spiritual/religious groups).

  PublisherOA PDFDOI

Frequent coauthors

• Nicola Dell

  Cornell University

  10 shared

• Thomas Ristenpart

  Cornell University

  9 shared

• Sam Havron

  Cornell University

  6 shared

• Rahul Chatterjee

  University of Wisconsin–Madison

  5 shared

• Deborah Estrin

  Cornell University

  4 shared

• Natalya N. Bazarova

  Cornell University

  3 shared

• Damon McCoy

  New York University

  3 shared

• Jackeline Palmer

  Hunter College

  3 shared

Labs

• Socially Responsible Computing @ Brown CSPI

  Develop and implement content on the social impact of technology in CS courses.