Survey identifies three proposed priority tools for agentic AI in research, ranked by the global research community.
The results from the global community consultation on agentic AI in research are now available. The consultation explored current use of agentic AI in research and gathered perspectives on its value throughout the research lifecycle.
Launched in November 2025, the consultation defined agentic AI as ‘artificial intelligence systems capable of autonomous operation with minimal human oversight.’ It was open to all, regardless of AI expertise or experience, or geographical location. The consultation builds upon RDA’s previous collaboration with Microsoft, which identified agentic AI as a capability need amongst researchers and recommended investment in automated data preparation tools.
Methodology and Participation
The consultation comprised two components: four online information sessions held across different time zones and an anonymous 15-minute survey. Survey respondents (n=83) evaluated 11 proposed agentic AI tools that spanned the research lifecycle, from planning and funding through to publication and impact reporting.
European respondents (n=44) represented the largest regional cohort and researchers and scientists (n=33) comprised the largest stakeholder group.
Results
Three proposed tools emerged as clear community priorities:
- The Literature Librarian, which would search literature using natural language queries integrated with library subscriptions;
- The Data Director, designed to support research data preparation and sharing in compliance with FAIR principles; and
- The Funding Finder, which would identify relevant funding opportunities and support application processes.
These top three rankings were reflected across both regional and stakeholder analyses, with the Literature Librarian, Data Director, and Funding Finder consistently appearing amongst the highest priorities.
Overall, participants portrayed a complex, ambivalent future regarding agentic AI in research.
Citation:
Clare, C., Collinson, M., Hanahoe, H., Payton, R., & Sharma, C. J. M. (2026). Global Community Priorities for Agentic AI in Research. Community Consultation Results. Zenodo. https://doi.org/10.15497/RDA00151
Next Steps
Insights from this consultation will guide the next phase of work in 2026, which intends to focus on collaborative community development of an open, technology-agnostic blueprint for a priority agentic AI tool. Any future work would be executed within the RDA framework and adhere to RDA Guiding Principles.
Contact
For general enquiries contact the RDA Secretariat ([email protected]). If you’re interested in being kept informed about potential future initiatives in this area or would like to get involved in the next phase of this project, please contact Connie Clare, Community Development Manager, at [email protected].
Comments
No comments found.