Schedule
8:30 - 9:00 Asia Biega
9:00 - 9:30 Krishnaram Kenthapadi
9:30 - 10:00 Negar Romstamzadeh
10:15 - 10:45 Break
10:45 - 12:00 KDD Keynote
12:00 - 1:00 Break
1:00 - 2:00 Panel
2:00 - 3:00 Live AI Ethics Debate on Accountability
3:00 - 3:30 Break
3:30 - 4:30 Three Minute Thesis Competition
4:30 - 5:30 Paper Presentations
From Regulations to Implementations: The Role of Research in Effective AI Governance in the Public Interest
Abstract: Policies and regulations are some of the pillars of Responsible AI. Their goals are broad: from user data protection, through securing the rights of training data contributors, to addressing broader considerations such as protecting society from systemic risks. Yet, despite the growing number of such regulations, the practical implementation of their underlying requirements in AI systems often falls short. This talk will explore the computational and organizational factors contributing to these shortcomings and discuss how researchers can contribute to improved governance of AI technologies in the public interest.
Dr. Asia Biega is a computer scientist and a tenure-track faculty member at the Max Planck Institute for Security and Privacy (MPI-SP) where she leads the interdisciplinary Responsible Computing group. Her research explores the intersection of computing and society, with a focus on tech governance & regulation, social responsibility, and digital well-being. She is a principal investigator in the CASA Cluster of Excellence and the FINDHR consortium, and has collaborated widely across disciplines and sectors. She has served as an external expert for the European Commission, a Steering Committee member of ACM FAccT, as well as the General Co-Chair of FAccT 2025, the biggest academic conference focusing on the societal impacts of computing. Her work has been recognized with Council of Europe’s Rodota Award for innovative research in data protection, SaTML Notable Reviewer Award, the GI-DBIS Dissertation Award of the German Informatics Society, and she has been named one of the 2025 "100 Brilliant Women in AI Ethics".
Trustworthy Health AI: Challenges & Lessons Learned
Abstract: While generative AI models and applications have huge potential across healthcare, their successful deployment requires addressing several ethical, trustworthiness, and safety considerations. These concerns include domain‑specific evaluation, hallucinations, truthfulness and grounding, safety and alignment, bias and fairness, robustness and security, privacy and unlearning, calibration and confidence, and transparency. In this talk, we first highlight the key challenges and opportunities for AI in healthcare, and then discuss unique challenges associated with trustworthy deployment of generative AI in healthcare. Focusing on the clinical documentation use case, we present practical guidelines for applying responsible AI techniques effectively and discuss lessons learned from deploying responsible AI approaches for generative AI applications in practice. In particular, we present insights from building and deploying AI agents as part of Oracle Health Clinical AI Agent.
Krishnaram Kenthapadi is the Chief Scientist, Healthcare Agents and AI Platform at Oracle, where he leads the AI initiatives for Clinical AI Agent and other Oracle Health products. Previously, he led AI safety, trustworthiness, and responsible AI initiatives at Fiddler AI, Amazon AWS AI, and LinkedIn, and served as LinkedIn’s representative in Microsoft’s AI and Ethics in Engineering and Research (AETHER) Advisory Board. Prior to that, he was a Researcher at Microsoft Research Silicon Valley Lab. Krishnaram obtained his Ph.D. in Computer Science from Stanford University in 2006. He has published 60+ papers, with 7000+ citations and filed 150+ patents, 72 of which have been granted. He has given invited talks and tutorials at leading research conferences and industry forums, and received research recognition awards.
Dr. Negar Rostamzadeh is a Staff Research Scientist on Google’s Responsible AI team, where she focuses on evaluating machine learning systems and their social impact. Her research spans computer vision, multimodal learning, and video understanding. Previously, she was a Research Scientist at Element AI, and earned her PhD from the University of Trento, with internships at MILA, Google, and Queen Mary University of London. Her work combines technical innovation with a deep commitment to ethical and responsible AI.
Dr. Elliot Creager is an Assistant Professor in Electrical and Computer Engineering at the University of Waterloo, and a Faculty Affiliate at both the Vector Institute and the Schwartz Reisman Institute. His research focuses on machine learning, with emphasis on algorithmic fairness, representation learning, and robustness. He earned his PhD from the University of Toronto, where he also conducted research at Google Brain.
Dr. Benjamin Fung is a Full Professor in the School of Information Studies at McGill University, Canada Research Chair in Data Mining for Cybersecurity, and an Associate Member of the School of Computer Science. His research spans data mining, machine learning, privacy, cybersecurity, and building engineering, with over 180 publications and 17,000+ citations. He leads the McGill Data Mining and Security Lab and serves as Editor-in-Chief of Sustainable Cities and Society: Advances. Dr. Fung’s work has influenced both academia and industry, with applications in crime investigation and authorship analysis reported worldwide.
Dr. Anastasia Kuzminykh is an Assistant Professor at the University of Toronto in the Faculty of Information, cross-appointed with the Department of Computer Science. She is a member of The Data Science Institute, a Faculty Affiliate with The Schwartz Reisman Institute for Technology and Society, the Director of the Toronto Human-AI Interaction Research School (THAI RS), and the Director of the COoKIE AI research group. Her work is situated at the intersection of AI Agent Design, Explainable AI, AI Safety, Natural Language Processing, and Conversational Interface Design, touching on diverse technologies, including LLMs, generative AI systems, and AI-based decision support systems to advance our knowledge on different facets of Human-AI Communication. Dr. Kuzminykh received her Ph.D. in Computer Science from the University of Waterloo and her background incorporates experiences in Human-Computer Interaction, Cognitive and Neuropsychology, Ethnographic Research, Information Science, and Systems Design, which allows her to combine diverse theoretical and practical approaches in her research program. Dr. Kuzminykh actively collaborates with industry and academia worldwide, her work is internationally recognized through awards and extensive funding, continuously published in the premier ACM venues, and gets broad coverage on TV, radio, and online media.
Dr. Atrisha Sarkar is an Assistant Professor in Electrical and Computer Engineering at Western University and heads the Humans and Autonomous Agents Lab. She is also a Faculty Affiliate at the Schwartz Reisman Institute for Technology and Society. Her research focuses on human-centric multiagent systems, integrating behavioral game theory, software engineering, and human-AI interaction to design safe, reliable AI systems that prioritize user and societal well-being.
Abstract: Artificial intelligence is facing a crisis. Humans are consuming far too many resources, impeding the progress of larger, faster, newer models. It’s time to drink less, shower less, and to prioritize AI.
Recent developments in AI technologies have relied on scaling technical infrastructure and capital. While improvements in benchmark performance have been put under the spotlight, environmental and social harms are obfuscated by Big Tech developing the models. Through an art-based initiative, titled Save the AI, we lean into satire and absurdity to visceralize AI’s staggering environmental and social impact. We designed posters, stickers, and postcards to be placed in public spaces to connect to people's basic needs for resources that are fundamental to their survival and AI's increasing demand for these same resources.
Presenter Bio: Han Qiao is a PhD student at the Faculty of Information, University of Toronto. Her research interests lie in the intersection of human-computer interaction, design and urban planning. Her recent projects explore the design of data visualization and exploration tools in the context of imagining sustainable futures.
Abstract: Access to historical clinical trial data is crucial to drive innovation, optimize trial design, and improve trial success. However, access to patient data is often restricted due to concerns over patient and sponsor privacy. Advancements in synthetic data generation provide a promising solution through generated datasets that maintain the statistical properties and insights of original clinical trials while safeguarding patient privacy. While most synthetic clinical trial data generation methods focus on preserving patient privacy, protection of trial sponsor confidentiality is overlooked. Systems that safeguard the sponsor identity are critical to enable and encourage data sharing. This study introduces a generalizable framework to enhance trial sponsor privacy in synthetic clinical trial data generation where the sponsor reidentification accuracy can result in up to 77% reduction. The proposed framework is agnostic to the underlying synthetic data generator model and emphasizes the importance of preserving sponsor privacy alongside patient confidentiality.
Presenter Bio: Dr. Chao Sang is a Senior Data Scientist at Medidata Solutions, where he focuses on the development of AI methodologies for healthcare applications, with a particular emphasis on clinical trial optimization, clinical predictive modeling, and privacy-preserving synthetic data generation. He holds a Ph.D. from the University of Pittsburgh with a background in computational modeling and applied biomedical research.
Abstract: As AI assistants become increasingly prevalent, their internal biases have a profound impact on the visibility and portrayal of brands and services. To enable systematic investigation of this phenomenon, we create a large-scale dataset consisting of user-chat simulations based on established consumer clusters. This framework allows researchers to rigorously examine whether AI models exhibit systematic preferences when responding to open-ended queries about commercial, government, and cultural entities. We demonstrate the framework's utility by comparing AI-generated rankings from ChatGPT-4o and Google Gemini 1.5-flash to real-world market data, measuring the economic and cultural impact of these biases and highlighting the importance of aligning AI systems with objective representation and diversity.
Presenter Bio: Jasmine Rienecker is a Senior AI Engineer and Researcher at creative and engineering agency Stupid Human. Her research focuses on how large language model bias impacts brands and services. Jasmine has a background building machine learning driven risk models at Goldman Sachs, as well as designing production-grade deep learning models for energy forecasting. She holds a Master’s degree in Mathematics and Computer Science from the University of Oxford, where she specialised in Explainable AI.
Abstract: There is increasing demand for accountability and auditing of ML and AI decision-making systems to ensure safety, security, and legal compliance. Issues may stem from outdated training data, model flaws, or insufficient safeguards during development. This talk will overview a novel system (AVOIR) designed to audit ML and AI models for regulatory fairness, including a custom domain specification language for specifying regulatory fairness requirements, and efficiently leveraging recent theoretical advancements in confidence sequence sets and conformal prediction.
This is joint work with P. Maneriker (OSU/Dolby Laboratories), A. Payani (Cisco Research), A. Srinivasan (OSU/Georgia Tech), A. Vadlamani (OSU).
Presenter Bio: Dr. Srinivasan Parthasarathy received his PhD in Computer Science from the University of Rochester, New York, USA. He is a Professor and University Distinguished Scholar in the Computer Science and Engineering Department at the Ohio State University (OSU). He directs the data mining research laboratory at OSU. His research interests are broadly in the areas of Data Mining, Databases, Bioinformatics, and High-Performance Computing. He is a recipient of an Ameritech Faculty fellowship in 2001, an NSF CAREER award in 2003, a DOE Early Career Award in 2004, and multiple fellowships from IBM, Google, and Microsoft. His papers have received seventeen best paper awards or similar honors from leading conferences in the field, including ones at SIAM international conference on data mining (SDM), IEEE international conference on data mining (ICDM), the Very Large Databases Conference (VLDB) ACM Knowledge Discovery and Data Mining (SIGKDD), ACM Web Search and Data Mining, The Web Conference, the ACM Bioinformatics Conference and the Intelligent Systems and Molecular Biology conference. Many of his works have transitioned to practice (commercial implementations of Eclat, MLR-MCL, and graph sparsification are used by many companies; and his work on the use of Zernike modeling for the early detection of Keratoconus is widely used in clinics). He serves or has served as one of the chairs of leading conferences in the fields of data mining, databases, and high-performance computing. He currently serves on the editorial boards of several journals, including Data Mining and Knowledge Discovery: An International Journal. He completed his final term as chair (elected) of the steering committee for the SIAM data mining conference series. He is a Fellow of the IEEE, the Risk Institute, the AAIA, and the Robert Bosch Center for Data Science and AI.