We recently had our monthly AI In Research Community Meet Up. Thanks for all who joined the conversation, the latest meet raised key topics such as training frameworks for researchers, ethical considerations, licensing restrictions, and the integration of AI tools into research workflows.
The community shared insights on developing tailored training programmes for researchers, negotiating licence terms with resource providers, ensuring transparency in research methods using AI tools, and adapting institutional policies to keep pace with advancements in AI technologies. There was a consistent theme of the importance of collaboration amongst universities to address shared challenges in this rapidly evolving space.
Key takeaways
- AI Training for Researchers and Academics: The discussion highlighted the need for tailored training programmes to help researchers and academics use AI tools effectively. Participants emphasised the importance of integrating AI topics into existing training sessions, such as research integrity or postgraduate researcher courses, to reach a wider audience.
- Challenges in Training Engagement: It was noted that mandatory training sessions often attract participants who already have knowledge of the subject matter while missing those who may benefit most. To address this issue, blending AI topics into broader training sessions was suggested as a strategy to increase engagement.
- Lack of Awareness About AI Tools: Concerns were raised about researchers’ limited knowledge of existing AI tools that could assist with tasks like literature reviews and grant writing. It was suggested sharing best practices and conducting sessions to introduce researchers to lesser-known tools like Illicit, Scite, and Scholarcy.
- Ethical Concerns with AI Tools: Discussions revealed ethical dilemmas surrounding the use of AI in research workflows. The community also expressed concerns about the burden placed on researchers to ensure compliance with ethical guidelines while using AI tools. This highlighted the need for clear frameworks to guide researchers in using AI responsibly without adding excessive workload.
- Licensing Restrictions on AI Usage: Some electronic resource providers are adding restrictive terms to their licences that prohibit the use of their content with AI tools. This poses challenges for universities that have integrated AI tools like Microsoft Copilot into their systems. People in the sector are making efforts to negotiate licence terms with electronic resource providers to allow secure use of AI tools within university systems. However, there remains resistance from providers and there could be a need for universities to collectively push back against restrictive licensing practices.
- Data Security Risks with AI Tools: Participants discussed risks associated with data security when using AI tools like Copilot. Issues were highlighted with unauthenticated versions of Copilot and its ability to access sensitive data unintentionally through web browsers like Edge. The importance of using licensed versions of Copilot to protect sensitive or confidential research data was also stressed.
- Transparency in Research Methods Using AI Tools: The importance of researchers selecting appropriate tools for their work and ensuring transparency in their methods was discussed. Ethical AI in research should start from being open about use and also using tools which meet standards.
- Emerging Trends in AI Tool Development: Discussions touched on recent developments in AI tools, including changes in security guarantees for Copilot and new measures to address copyright issues in creative outputs generated by generative AI platforms. Participants acknowledged the fast-evolving nature of the space and the need for continuous updates to policies and training materials.
Going forwards:
- Researchers are strongly advised to use only licensed versions of Copilot when handling confidential or commercially sensitive information to ensure data protection and avoid disclosure risks.
- Universities should aim to integrate discussions about AI tools into existing mandatory training sessions (e.g., research integrity training) rather than offering separate courses, as this approach is more likely to engage researchers who may not voluntarily attend standalone sessions.
- There is an upcoming AIRON panel discussion on the 26th of February at 4pm on AI in research assessment. As we look ahead to REF 2029, this conversation raises an important and shared question for the sector: How can AI help reduce burden while preserving trust, transparency, and equity in research assessment?
- The AI.RDN+ project formally kicked off with a successful launch conference, bringing together around 60 participants from over 20 universities and industry partners to set the stage for this exciting new initiative. The event featured keynote insights on the evolving role of AI in doctoral research from academic leaders and practitioners and marked the beginning of a collaborative effort to explore and support the responsible use of AI tools across the doctoral research ecosystem.
- The next Jisc AI in Research community meet-up will be taking place this month on February the 24th at 3pm. Event details and a reminder will be sent out to members closer to the date.
Find out more by visiting our Artificial Intelligence page to explore publications and resources, learn more about our communities and sign up for our AI Literacy training.
For regular updates from the team sign up to our mailing list.
Get in touch with the team directly at AI@jisc.ac.uk