Large Language Models and the Disappearing Private Sphere
Organization
Queen’s University
Published
2024
Project Leader(s)
Dr. Catherine Stinson
Summary
Large language models have suddenly become a source of interest (or consternation) for teachers, writers, business leaders, general Internet users, AI researchers and policy makers. It remains to be seen whether these tools will revolutionize industries or gradually reveal themselves to be no more interesting than spell-checkers. In the meantime, they have stirred up a hornet’s nest of questions about privacy rights, copyright and research methodology. Researchers relied upon expertise in AI, research ethics, and technology policy to review the technical, ethical and policy implications of large language models.
This report begins with an accessible introduction to the technology and the regulatory context. Researchers then reported on a survey conducted with university research ethics experts across Canada about how research ethics review boards are currently handling AI research and web scraping, and how they think current practices might need to change. Researchers then conducted a literature review of the technical literature about privacy leakage from large language models, and supplementary experiments filled in some gaps in existing research. Finally, researchers discussed the gaps in Tri-Council policies concerning artificial intelligence research, and created a series of recommendations for the Tri-Council and for institutional research ethics boards. Then discussed remaining gaps in technical work on mitigating privacy leakage, and provided recommendations for artificial intelligence developers and users of large language models. Finally, researchers made recommendations for policy makers that built upon previous work by the Office of the Privacy Commissioner of Canada.
Project deliverables are available in the following language(s):
English
OPC Funded Project
This project received funding support through the Office of the Privacy Commissioner of Canada’s Contributions Program. The opinions expressed in the summary and report(s) are those of the authors and do not necessarily reflect those of the Office of the Privacy Commissioner of Canada. Summaries have been provided by the project authors. Please note that the projects appear in their language of origin.
Contact Information
Dr. Catherine Stinson
Queen’s National Scholar in Philosophical Implications of Artificial Intelligence
Assistant Professor, Philosophy Department & School of Computing
Queen’s University
Email: c.stinson@queensu.ca
- Date modified: