RobustifAI Workshop
Robustifying Generative AI for Reliable, Safe, and Human-Centric Systems

Important dates (all times are 23:59 Anywhere on Earth, UTC-12)
Submission: 8 May 2026 (link_to_submission)

Notification: 8 June 2026

Workshop: 1-day workshop within 15-17 August 2026 (at IJCAI/ECAI 2026)

Workshop overview

Robustifying Generative AI (GenAI) addresses the challenge of making generative AI systems reliable, safe, and effective in real-world deployment. The need for robustness becomes increasingly important as foundation models and GenAI systems are integrated into critical and high-impact applications, raising questions such as: how should robust GenAI systems be designed across model, deployment, and user interaction layers? What kinds of failures should they be able to detect, tolerate, or prevent under distribution shift, adversarial inputs, tool use, and human interaction? How should monitoring, governance, and assurance mechanisms be incorporated into the lifecycle of these systems? And how should robustness be evaluated in ways that go beyond narrow benchmark performance? 

The RobustifAI workshop at IJCAI-ECAI provides a forum for discussing recent research on robustifying generative AI systems, highlighting promising approaches, and encouraging further work that connects technical, operational, and human-centered perspectives. The workshop aims to bring together researchers from machine learning, verification and testing, security, neural-symbolic AI, and human-centered AI to discuss methods, evaluation protocols, and assurance practices for GenAI robustness, with particular emphasis on demanding deployment contexts such as human cyber-physical systems. 

In addition to encouraging descriptions of original or recent contributions on GenAI robustness, including theory, algorithms, tools, empirical studies, demonstrations, and deployment experiences, we also welcome contributions that survey related work; identify key open problems requiring further research; or highlight practical and scientific challenges relevant to the broader AI community, together with possible paths toward addressing them. 

Format

The workshop will comprise several events:

• keynote talks;

• presentation of papers accepted into the workshop;

• poster presentations; and 

• a discussion panel for participants to discuss key ideas, gaps, and future directions.

Contact

If you have any questions about the IJCAI Workshop on RobustifAI, please contact the organizers via the following email

chih-hong.cheng AT uol.de