Co-located with IEEE MASS 2025 | October 6–8, 2025
As smart systems integrate generative AI with edge-cloud architectures, their security paradigms face unprecedented challenges. Data security and LLM safety are crucial for trustworthy smart infrastructure. Data security protects sensitive information throughout its lifecycle, while LLM safety ensures the reliability and ethical alignment of AI behaviors. These two are intertwined: compromised data integrity directly undermines model robustness, and unsafe LLM outputs can lead to data breaches in interconnected smart ecosystems. This is especially critical in mobile environments with dynamic connectivity, resource constraints, and device diversity. Traditional frameworks, developed for static systems, are inadequate to address these dynamic, resource-constrained scenarios. The limitation is increasingly evident in real-world incidents across healthcare, transportation, and smart manufacturing sectors.
This workshop aims to unify data protection and AI safety for next-generation smart systems. While the main conference covers broader mobile ad-hoc network security, we specifically focus on emerging threats arising from the fusion of generative AI and distributed intelligence. Topics will span novel threat models, architectural safeguards, and evaluation methodologies that jointly strengthen data flows and AI behaviors. The outcomes will deliver practical guidelines to help academia and industry harness LLMs' transformative potential while mitigating systemic risks in critical applications like smart healthcare, autonomous transportation, and industrial IoT.
We invite submissions on emerging challenges including but not limited to:
Format: All submissions should be written in English with a maximum length of 6 single-spaced, double-column pages using 10pt fonts on 8.5 in x 11 in paper, including all figures, tables, and references, in PDF format. Authors must use the Manuscript Templates for IEEE Conference Proceedings.
Review: Reviewing will be single-blind, i.e., authors can keep their names on their submitted workshop paper.
Submission Portal: Click here.
Authors are invited to submit original, unpublished workshop papers that are not currently under review elsewhere. Accepted workshop papers will be included in the conference proceeding published in the IEEE Xplore Digital Library. For all workshop papers, IEEE reserves the right to exclude the workshop paper from distribution after the conference if the workshop paper is not presented at the conference.