AI4SafeDist 2026

Artificial Intelligence (AI) components are increasingly integrated into safety-critical systems across domains such as transportation, industrialautomation, healthcare, and smart infrastructures. These systems are progressively distributed, interconnected, and data-driven, combining AI-based decision-making with complex software and hardware ecosystems. While AI technologies provide new capabilities, they also introduce significant challenges regarding safety assurance, reliability, security, and dependability assessment. Traditional engineering methodologies for safety-critical systems are often not directly applicable to adaptive and data-driven components, especially when deployed in distributed and heterogeneous environments. Recent editions of SafeComp have highlighted growing interest in AI-based systems, safety assurance, runtime monitoring, and certification challenges. However, systematic approaches for assuring AI-enabled distributed systems remain an open research challenge.

AI4SafeDist 2026 aims to provide a focused forum within SafeComp for discussing methods, tools, frameworks, and practical experiences addressing the safety and security assurance of AI-enabled distributed systems.

In particular, the workshop emphasizes the integration of safety, security, and dependability assurance approaches for AI-enabled distributed systems, directly aligned with the core topics of the SafeComp conference. The workshop targets researchers, practitioners, and industry experts working at the intersection of AI, dependable systems, cybersecurity, and safety engineering.

Goals:

• Foster interaction between AI researchers and safety engineering communities.

• Identify challenges in assuring and certifying AI-enabled safety-critical systems.

• Promote methodologies for safe and secure deployment of distributed AI.

• Encourage discussion of early-stage ideas, work-in-progress results, and industrial experiences.

• Contribute to shaping research directions at the intersection of AI, safety, and security.

Topics of interest include, but are not limited to:

• Safety and reliability in distributed and AI-enabled systems

• Safety/security co-engineering in AI-enabled architectures

• Privacy-preserving and secure machine learning (e.g., federated learning, secure computation)

• Robustness, explainability, trustworthiness, and resilience of AI systems

• Fault tolerance and dependable AI-enabled architectures

• Model-based and data-driven dependability assessment

• Edge-to-cloud and cyber-physical AI systems in regulated environments

• Multi-concern dependability assurance combining safety, security, and AI

• Socio-technical, regulatory, and ethical aspects of safe and secure AI

• Industrial case studies involving AI in safety-critical environment

• TDB

TBD

• Manuel F. Dolz, Universitat Jaume I, Spain, dolzm@uji.es

• Sandra Catalán, Universitat Jaume I, Spain, catalans@uji.es

• Larbi Boubchir (University of Paris, France)

• Damien Ligier (DESILO, Seoul, South Korea)

• Christian Prehofer (Technical University of Munich, Germany)

• Bernardo Pulido-Gaytan (National College of Ireland, Ireland)

• Leonel Sousa (University of Lisbon, Portugal)

All papers will be reviewed by at least three reviewers. Workshop proceedings will be provided as complementary book to the SafeComp Proceedings in Springer LNCS. Papers (6-12 pages) will be reviewed by at least three reviewers. Please keep your paper format according to Springer LNCS style guidelines.

Submission will be via EasyChair: TDB

Deadlines:

• Full paper submission: 4 May 2026

• Notification of acceptance: 18 May 2026

• Camera-ready submission: 08 June 2026

• Workshop: 22 September 2026

Information about registration at SAFECOMP 2026 website.