ICANN & PAVE LabsUniversity of South Carolina
Human-centered lab · launching 2026

PAVE Lab

People, AI, Values & Evaluation

The PAVE Lab studies, designs, and evaluates artificial intelligence systems with a focus on the people who interact with them and the values that shape their design and use. Our research spans human-AI interaction, conversational and generative AI, responsible and ethical AI, and applied AI. We combine computational, experimental, and human-centered methods to understand how people communicate, seek information, collaborate, and make decisions with AI, while developing methods to evaluate these systems for effectiveness, reliability, bias, trust, and accountability.

Human-AI interactionConversational & generative AIResponsible AIAI evaluationApplied AI

About the lab

PAVE launches at the School of Information Science, University of South Carolina, in 2026. It is the human-centered counterpart to the ICANN Lab, which builds and probes the systems themselves. The two labs share students, methods, and most of their interesting questions.

The lab works across disciplines by design. Questions about how AI reshapes reference work, records management, health information, or classroom practice cannot be answered from inside computer science alone, and they cannot be answered without it either. Our longer-term aim is to formalize PAVE as a university center.

Focus areas

  1. Human-AI interaction and information seeking

    How people ask, follow up, and decide when the system talks back. We study the shift from queries to conversations, what people expect a generative AI system to know, and how those expectations shape the questions they are willing to ask.

  2. Responsible and ethical AI

    Audits of demographic and stereotype bias in model behavior, with attention to consequential settings such as law enforcement and healthcare, where a small shift in a model's judgment lands on a real person.

  3. AI evaluation, trust, and accountability

    Methods for judging whether a system is reliable, whether a user's trust in it is calibrated to that reliability, and who answers for the system when it is wrong.

  4. Applied AI in public-interest settings

    Libraries, archives and recordkeeping, health and end-of-life information, reference services, and customer-facing work. Sectors where AI arrives with little institutional support and unusually high stakes.

  5. AI literacy, governance, and the information workforce

    Responsible computing education, professional competencies for information workers, and governance frameworks for institutions adopting AI-mediated services.

Ongoing projects

  • Auditing bias in model judgments

    Scenario-based testing of how demographic details change the decisions large language models recommend in high-stakes domains.

    Accepted, ASIS&T 2026

  • Human-AI collaboration in qualitative analysis

    Where model-assisted coding of health-related video content matches trained human analysts, and where it fails in ways that would change a study's conclusions.

    Funded

  • Bias and stereotype in generated dialogue

    What conversational systems reproduce when they generate open-ended text about people. Developed jointly with ICANN.

    In progress

  • InterPARES Trust AI

    AI for retention and disposition in trusted digital recordkeeping repositories, and assessment of personal information in records. AI consultant and co-applicant.

    SSHRC · through 2026

  • CIRCLE

    Cross-Campus Interdisciplinary Responsible Computing Learning Experiences: embedding ethics into computing coursework across multiple institutions.

    Mozilla Foundation · co-PI

  • Technology, governance, and hope

    An international program on how societies steer the technologies they adopt.

    Gilman Program · co-PI

  • AI and social justice

    A book on challenges and opportunities where AI meets questions of justice and access, under contract with Emerald.

    In progress

People

Dr. Souvick “Vic” Ghosh

Director and Principal Investigator · profile

Open positions

Doctoral Researcher

Open, Fall 2026

Graduate Research Assistant

Open, Fall 2026

Affiliated Faculty

Cross-campus collaborators

Undergraduate Researcher

Open, by semester

Student collaborators are credited as co-authors in the publications below. See the join page for how to apply.

Publications

  • Probability and prejudice: a scenario-based audit of demographic bias in LLM judgments across law enforcement and healthcare

    Ghosh, S., Mehra, B., & Lu, K. (2026). Proceedings of ASIS&T, 63. Accepted.

  • Exploring stereotype bias in large language model dialogue generation

    Ghosh, S., Dubin, R., & Charette, C. (2026). Proceedings of ASIS&T, 63. Accepted.

  • Scrolling for comfort: human-information interaction around palliative care on TikTok

    Ghosh, S., Du, T., & Dubin, R. (2026). Proceedings of ASIS&T, 63. Accepted.

  • Embedding responsible computing in an undergraduate computer vision course

    Attar, N., Ghosh, S., Heller, P., Villagran, M. A., & Hofman, D. (2026). AIxHEART. Accepted.

  • Assessing artificial intelligence and social justice intersections: challenges and opportunities

    Ghosh, S., Fu, H., Hofman, D., & Mehra, B. (2026). Advances in Librarianship, Emerald. Book, in progress.

  • Governing and designing AI-mediated library services

    Ghosh, S. (2026). In Library and Information Science Education around the World, IFLA Series. Book chapter, in progress.

  • From queries to conversations: examining human–GenAI information-seeking through Belkin's cognitive communication model

    Charette, C., & Ghosh, S. (2025). Proceedings of ASIS&T, 62(1), 105–116.

  • Human-AI collaborative content analysis: the efficacy and challenges of LLM-assisted content analysis for TikTok videos on palliative care

    Ghosh, S., Malempati, K., & Charette, C. (2025). Proceedings of ASIS&T, 62(1), 229–240.

  • Evolution of reference services in the era of generative artificial intelligence

    Aguiñaga, J., Mooradian, N., Ghosh, S., & Hofman, D. (2025). Proceedings of ASIS&T, 62(1), 13–25.

  • Tracing the past, predicting the future: a systematic review of AI in archival science

    Shinde, G., Kirstein, T., Ghosh, S., & Franks, P. (2025). Proceedings of ASIS&T, 62(1), 659–671.

  • Exploring dehumanization and infrahumanization as underlying factors in misinformation belief and spread

    Weiss, A., Ghosh, S., & Johnson, F. (2025). CAIS. Best Paper by a Practitioner.

  • Exploring self-dehumanization as a factor in misinformation belief and spread

    Weiss, A., Ghosh, S., & Johnson, F. (2025). ACM CHIIR, 326–332.

  • Empowering customer service with generative AI: enhancing agent performance while navigating challenges

    Costa, C., & Ghosh, S. (2025). Information Research, 30(iConf 2025), 150–158. DOI

  • Enhancing enrollment and participation in computing ethics through CIRCLE

    Attar, N., Ghosh, S., Villagran, M. A., & Hofman, D. (2025). IEEE EDUCON, 1–4.

  • Integrating AI into library systems: a perspective on applications and challenges

    Tai, I., & Ghosh, S. (2024). ACM/IEEE JCDL.

  • Looking ahead: incorporating AI in MLIS competencies

    Ghosh, S., & McCoy, D. (2024). School of Information Student Research Journal, 14(1), 3.

  • Decoding large language models: a systematic overview of socio-technical impacts, constraints, and emerging questions

    Kaya, Z. N., & Ghosh, S. (2024). arXiv preprint 2409.16974. arXiv

  • AI in archival science: a systematic review

    Shinde, G., Kirstein, T., Ghosh, S., & Franks, P. C. (2024). arXiv preprint 2410.09086. arXiv

  • Recommendations on using artificial intelligence in archival appraisal and selection

    Bayeck, R. Y., Colavizza, G., Bunn, J., Bell, M., & Ghosh, S. (2022). Comma, 2022(1), 219–225. DOI

Both labs, all years →

Join the lab

What I am looking for

PAVE is recruiting its first cohort for Fall 2026. I want students who are curious about what AI does to people and institutions, and who are willing to learn a method they do not already have. The most useful person on a project like this is often the one who notices that the question is wrong.

  • Backgrounds in information science, HCI, communication, public health, education, law, or computing
  • Interview and survey design, qualitative coding, or statistics, or the willingness to pick one up
  • The ability to write clearly about a messy problem, which counts for more than a specific toolchain

Email with the subject line PAVE Lab inquiry, a paragraph on a question about AI and people that you cannot stop thinking about, and your CV. Full details are on the join page.

souvick.ghosh@sc.edu