On April 8th, 2026, Temple University's Master of Public Policy program held a panel discussion on "Risks and Rewards of AI in the Public Sector." The discussion brought together experts working at the intersection of technology, policy, education, and journalism to explore how artificial intelligence is shaping government systems, services, and policymaking in Philadelphia, Pennsylvania, and across the country.
Panelists included:
- Dr. Joshua Mask, Assistant Professor of Economics - Instructional, Temple University
- Malcolm Burnley, Staff Writer, The Philadelphia Citizen and Philly Mag
- Dr. Sorelle Friedler, Shibulal Family Professor of Computer Science, Haverford College; Co-Director, Kim Institute for Ethical Inquiry and Leadership at Haverford College; Chair, U.S. Technology Policy Committee, Association for Computing Machinery; Nonresident Senior Fellow, The Brookings Institution
The event was moderated by MPP students Emanuel Schorsch and Jamya Johnson.
The discussion opened with panelists sharing their experiences working with and writing about AI before diving into real-world examples of its use in the public sector. Examples included SEPTA identifying vehicles blocking bus lanes, California piloting automated road and weather danger alerts, and improving processing requests at the U.S. Department of Veterans Affairs. The panel highlighted how AI is already being put to work in government, with promising results in areas like benefits delivery, traffic safety, and economic forecasting.
Panelists emphasized that not all AI is the same, and distinguishing between tools like ChatGPT and more specialized, trainable systems is critical when thinking about their use in government.
Additionally, panelists discussed the importance of human judgment, particularly in high-stakes decisions that affect people's lives. They emphasized that AI should support, not replace, human judgment as AI cannot “care,” making human oversight essential, especially in areas such as healthcare and public safety.
Another important point raised was the role of education and adaptation. As AI becomes more integrated into everyday work, panelists noted that students and professionals alike will need to learn how to use these tools effectively while still maintaining strong critical thinking and decision-making skills.
Furthermore, panelists addressed concerns around bias, transparency, and civil liberties, including the use of AI tools by immigration enforcement and the spread of AI-generated disinformation during local elections. The discussion also touched on the current policy landscape, noting a lack of consistency in how AI is regulated across states. Examples included local governments implementing AI use disclosures and Maine’s 2024 ban on certain uses of ChatGPT by executive employees. Panelists emphasized that residents should play a key role in shaping how AI is used by demanding transparency and accountability.
Dr. Friedler closed with a similar sentiment expressed by the other two panelists at the conclusion of the event, reminding attendees that “it’s not going to be a one-size-fits-all. Be skeptical. The story is being written by us, the users and constituents”.