Background: The landscape of artificial intelligence (AI) in rheumatology has evolved dramatically since 2020. Over 5,000 scholarly articles on AI in rheumatology were published by early 2026, with vertical growth in generative AI applications since 2025. AI has demonstrated remarkable capabilities: machine learning achieving 94.7% sensitivity and 93.6% specificity for sacroiliitis identification on MRI, matching or exceeding senior radiologists’ accuracy; deep learning models showing promise in rheumatoid arthritis flare prediction; and large language models being evaluated for clinical decision support. Recent German studies demonstrate AI scribes benefit both clinicians and patients. However, rapid AI advancement, emergence of large language models, the EU AI Act regulatory framework, and heightened focus on patient privacy necessitate updated guidance beyond the 2020 EULAR Points to Consider for big data in RMDs.

Objectives: To update and expand the 2020 EULAR Points to Consider for big data in RMDs to comprehensively address AI applications, incorporating regulatory frameworks (EU AI Act, Data Act, Data Governance Act), enhanced data privacy protections, development standards for AI-powered clinical decision aids, and requirements for explainability.

Methods: Following EULAR standardized operating procedures, a multidisciplinary international task force has been established with rheumatologists, AI specialists, data scientists, patient research partners, lawyers, and ethicists from >11 countries. A comprehensive literature review is being conducted covering MEDLINE, EMBASE, and Web of Science, focusing on AI and big data publications in RMDs since 2020, implementation studies, and healthcare AI ethics. Targeted searches of regulatory documents including the EU AI Act and Medical Device Regulation will be performed. The steering group will review evidence and formulate updated points to consider or recommendations addressing: machine learning for big data analysis, computer vision in medical imaging, natural language processing and large language models, diagnostic and prognostic decision-support systems, patient information and education, and ethical challenges. Consensus will be achieved through structured discussion and voting, with levels of agreement collected on a 0-10 scale.

Results: Results from this task force will be completed and presented at EULAR 2026.

Conclusions: This updated guidance aims to position EULAR at the forefront of responsible AI implementation, providing the rheumatology community with clinically relevant, implementable recommendations for navigating the evolving AI landscape while maintaining the highest standards of patient care. The framework seeks to balance innovation with safety, standardizing approaches to manage demand surges effectively in rheumatology practice.

REFERENCES: [1] Gossec L, Kedra J, Servy H, Pandit A, Stones S, Berenbaum F, Finckh A, Baraliakos X, Stamm TA, Gomez-Cabrero D, Pristipino C, Choquet R, Burmester GR, Radstake TRDJ. EULAR points to consider for the use of big data in rheumatic and musculoskeletal diseases. Ann Rheum Dis. 2020 Jan;79(1):69-76.

[2] Kedra J, Radstake T, Pandit A, Baraliakos X, Berenbaum F, Finckh A, Fautrel B, Stamm TA, Gomez-Cabrero D, Pristipino C, Choquet R, Servy H, Stones S, Burmester G, Gossec L. Current status of use of big data and artificial intelligence in RMDs: a systematic literature review informing EULAR recommendations. RMD Open. 2019 Jul 18;5(2):e001004.

Acknowledgments: NIL.

Disclosure of Interests: Latika Gupta: None declared, Vincenzo Venerito: None declared, Laure Gossec: None declared, Thomas Hügle Abbvie, GSK, J&J USB, Eli Lilly, Atreon, Fresenius Kabi, Eli Lilly, Paco Welsing: None declared, Gerd R. Burmester: None declared.