The IBM Research Lab in Dublin was founded in 2011 and ever since has consistently delivered major innovations in the areas of AI, security, hybrid cloud and quantum computing. As the only IBM Research lab in the European Union, the mission of the Dublin Lab, in addition to pursuing cutting-edge research for the future of computing, is to cultivate close relationships with academic and industrial partners, be one of the premier locations for world-class researchers, to promote women in IT and science, and to help drive Europe’s innovation agenda. At present, we are more than 80 researchers, scientists, technologists, engineers, designers, and thinkers, inventing what’s next in computing.

IBM Research Europe – Dublin is proud to be a key partner in the SEARCH project funded by the Innovative Health Initiative, aimed at advancing synthetic data generation for healthcare. SEARCH brings together leading institutions across Europe to develop privacy-preserving, high-quality synthetic datasets that can accelerate medical research and innovation while safeguarding sensitive patient information.

IBM Research Core Contributions to SEARCH

Our involvement in SEARCH spans several critical work packages, with a focus on two foundational pillars: platform architecture design and AI robustness.

1. Designing the SEARCH Platform Architecture. IBM leads the design and development of the technical architecture that underpins the SEARCH platform. This work contributes to shaping a platform that is:
   • Modular: allowing for flexible integration of components and tools across different healthcare use cases.
   • Scalable: capable of supporting large volumes of synthetic data generation and analysis.
   • Interoperable: aligned with healthcare data standards, enabling seamless collaboration across institutions and systems.

   Our involvement helps support SEARCH serve as a reliable foundation for synthetic data innovation across Europe.

2. Evaluating Robustness of AI and Large Language Models (LLMs)In the Medical Data Generation work package, IBM focuses on evaluating and enhancing the performance and safety of large language models (LLMs) used to generate synthetic healthcare data. The team assesses how these models perform, identifies potential vulnerabilities, and looks at ways to reduce risks such as information leakage or bias.

   By combining architectural design with careful evaluation of AI robustness, IBM Research contributes to making the SEARCH platform both well-structured and fit for purpose in the sensitive domain of healthcare data. Through its role in SEARCH, IBM Research continues to demonstrate leadership in trustworthy AI and privacy-preserving technologies; helping to build a future where synthetic data can unlock new possibilities in healthcare without compromising the privacy of individuals.

IBM Research Europe – Dublin SEARCH Team: Anisa Halimi, Research Scientist