Accepted Papers & Talks
Papers
A Modular AI Testing Framework for Trustworthy AI: Proof-of-Concept Implementation
Maximilian Pintz, Daniel Becker and Michael Mock
Architectural Mitigation of Control AI Risk Factors for Safe Human-Robot-Collaboration
Andreas Kreutz, René Beck, Gereon Weiss and Satoshi Otsuka
Uncovering Unsafe Feature Interactions in Vehicle Control Using Generative AI and Digital Twins
Laure Millet, Justin Kernot, Arun Adiththan, Ramesh S, Rami Debouk and Jeffrey Joyce
AURORA Networks: Auto-associative Universal Real-time Outlier Risk Assessment Networks
Moritz Zink, Daniel Grimm and Eric Sax
Does not impute! Performance and ethical implications of missing data for an AI-based diabetes co-morbidity predictor
Philippa Ryan, Berk Ozturk, Laura Fearnley, Tom Lawton and Ibrahim Habli
Facilitating Fault Tree Analysis with Generative AI
Yujiao Shentu and Mario Trapp
Efficient Safety Retrofitting Against Jailbreaking for LLMs
Dario Garcia-Gasulla, Adrian Tormos, Anna Arias-Duart, Daniel Hinjos, Oscar Molina-Sedano, Ashwin Kumar Gururajan and Maria Eugenia Cardello
Risk Analysis of One-Pixel Image Defects in Safety-Critical Deep Neural Networks
Krystian Radlak, Adam Popowicz, Michal Szczepakiewicz and Paweł Zawistowski
Safe Adversarial Control Through Interaction
Benedikt Rank and Mario Trapp
Talks
Runtime Safety Engineering for AI Robotics: Lessons from Scaling Trust at Amazon
Gianpiero Negri
The BIG Safety Argument
Ibrahim Habli
Assuring Security of Agentic AI: Risks, Attacks, and System Context
Damir Safin, Tomas Bueno Momcilovic, and Dian Balta