Call for Papers

Important Dates

First submission round:

• Paper submission deadline: June 22, 2026 AoE
• Notification to the authors: July 22, 2026 AoE
• Camera-ready deadline: August 7, 2026 AoE

Second submission round:

• Paper submission deadline: July 20, 2026 AoE
• Notification to the authors: August 9, 2026 AoE
• Camera-ready deadline: August 16, 2026 AoE

Overview

RAISE, co-located with ESORICS 2026, aims to bring together researchers and practitioners from academia, industry, and public institutions to discuss recent advances and open challenges in the design, deployment, evaluation, and protection of AI-based cybersecurity systems. The workshop focuses on understanding the security properties, limitations, and failure modes of AI-based components as they are designed and operated in realistic security environments, not in isolation, and at scale.

RAISE welcomes contributions that critically analyze how AI-based cybersecurity mechanisms fail, degrade, or are exploited in practice, as well as work that explores how AI can be reliably and securely leveraged to strengthen cybersecurity systems under real-world operational constraints. Both defensive and offensive perspectives are within scope, provided that they are grounded in realistic threat models, deployment assumptions, and system constraints. The workshop particularly encourages submissions from industrial practitioners and applied researchers, whose insights are essential for understanding operational realities, system integration challenges, and real-world trade-offs. In addition to traditional research papers, we explicitly welcome experience reports, empirical studies, preliminary results, and negative or unexpected findings that shed light on the gap between academic assumptions and operational reality.

Topics of Interest

Topics of interest include but are not limited to:

• AI-driven cybersecurity systems under real-world operational constraints, such as latency budgets, resource limitations, limited data availability, and human-in-the-loop workflows, for applications like malware detection, fraud prevention, abuse detection, incident response, threat intelligence, forensics, and automated security workflows.
• Threat modeling and adversarial exposure of AI-based security components, focusing on how AI systems used for cybersecurity can be attacked, bypassed, poisoned, manipulated, or otherwise exploited once integrated into real-world systems or services.
• Evaluation methodologies under operational and adversarial constraints, focusing on metrics, protocols, benchmarks, and tools for assessing the robustness, privacy, reliability, and security of AI-based cybersecurity systems beyond controlled or benchmark-only settings.
• Trust, explainability, and compliance in deployed AI-based security systems, focusing on how trustworthiness requirements, such as explainability, accountability, and regulatory compliance (e.g., GDPR, NIST, EU AI Act), interact with the security of AI systems used in practice.
• Defensive engineering and operational hardening of AI-based security systems, focusing on secure integration practices, resilience mechanisms, monitoring, retraining, fallback strategies, and runtime safeguards for AI-based cybersecurity components.
• Human–AI interaction, data pipelines, and model lifecycle security, focusing on the role of human operators, feedback loops, data collection, logging and lifecycle management in the security of AI-driven cybersecurity systems.

Submission Guidelines

Submissions must be written in English and provided in PDF format, using the Springer LNCS template.

We invite submissions in the following categories:

• Full papers, up to 12 pages, presenting novel contributions, experimental studies, or in-depth case analyses that address the security of AI-based cybersecurity systems or the use of AI for cybersecurity in realistic settings. Suitable submissions may focus on system-level design, evaluation under realistic threat models, deployment experiences, operational constraints, or empirical insights gained from real-world or industrial environments. Papers presenting tools, datasets, benchmarks, or measurement infrastructures that support the realistic evaluation or deployment of AI-based cybersecurity systems are also welcome.
• Non-archival papers, up to 6 pages, excluding references and appendices, describing work in progress, preliminary findings, experience reports, or negative results related to practical and industrial challenges in deploying AI securely, or in deploying AI-driven security tools in operational environments. These submissions may discuss early results, unexpected failure modes, deployment failures, or emerging challenges, along with potential solution directions. We also welcome papers presenting practical implementations, open-source libraries, or tooling that support real-world security testing or deployment, such as libraries for red-teaming and auditing of AI-driven systems, or development frameworks for building and integrating AI-based cybersecurity applications.

Accepted full papers may be extended up to 16 pages in the camera-ready version to accommodate reviewer feedback and will be included in the Springer Lecture Notes in Computer Science (LNCS) workshop proceedings. Non-archival papers may include previously published or concurrently submitted work and will not be included in the proceedings.

Beyond methodological novelty, RAISE strongly encourages submissions that emphasize practical relevance, deployment considerations, reproducibility, and system integration. Shorter submissions are welcome and will not be penalized if they clearly convey valuable insights, lessons learned, or practical experience. The workshop places emphasis on substance and relevance rather than paper length.

All submissions will undergo a double-blind peer-review process and will be reviewed by at least two members of the Program Committee.

Submission link: https://easychair.org/my/conference?conf=esorics2026
(select the "Workshop on Real-world AI Security and Engineering for Cybersecurity Systems" track).