July 29, 2025—The Software Engineering Institute (SEI) Artificial Intelligence (AI) Division invites AI researchers and practitioners to submit contributions to its Engineering Safety-Critical AI Systems symposium. The event, part of the Association for the Advancement of Artificial Intelligence (AAAI) 2025 Fall Symposium Series, will be held in Arlington, Virginia, November 6-8. Paper submissions are due to the AAAI EasyChair site by August 11.

AI has increasing application in high-risk settings, but few foundational practices exist for engineering safe AI systems, and research in safety engineering for AI remains scattered across disparate fields of study. The symposium will address the fundamental question, How should we build AI systems for safety-critical applications?

The SEI invites paper submissions for one of two tracks: Research and Development and Case Studies in Engineering AI Systems. While most topics around safety and AI are welcome, symposium organizers are especially interested in topics on how to engineer AI systems now and in the future. Areas of interest include formal specification and verification of AI systems, software architectures for increased AI system safety, and human–AI interaction in safety-critical systems, among others.

“AI holds great promise for advancing capability, especially in safety-critical scenarios and applications,” said Eric Heim, the symposium’s organizer and the chief scientist of the SEI’s AI Division. “Those applications are the ones that require the most care and discipline in design, development, and deployment. Participants in our symposium will be advancing the community’s understanding of how we should be using, building, and engineering those systems.”

The symposium will feature a mix of keynote speakers, selected paper presentations, and panel discussions. The SEI is organizing the event in collaboration with researchers from CrowdStrike, MIT Lincoln Laboratory, Duke University, Johns Hopkins University, Carnegie Mellon University, the University of Virginia, the University of Maryland, and the University of California, Berkeley.

Learn more about the Engineering Safety-Critical AI Systems symposium, including author instructions, on the AAAI 2025 Fall Symposium Series website.