We had our latest community meet up and it was an engaging conversation. Thanks to our members for coming together to contribute. This month’s key discussions looked at the challenges related to the use of AI tools in academic research, issues surrounding licensing complexities, data privacy risks associated with free tools, and the need for institutional policies.
We also explored strategies for creating responsible researcher behaviour through training, policy development, and thinking about research integrity frameworks. The meet up ended with a call for greater collaboration within the community to share resources and insights while advocating for national-level solutions to address the common challenges.
Key takeaways
Challenges with AI Tools in Research: Researchers are beginning to develop their own machine learning tools due to financial constraints and limited access to advanced AI platforms like ChatGPT, CoPilot, and Gemini. This raises concerns about data security and the need for guidance on safe tool usage.
Concerns About Free Tools: Free AI tools often come with significant risks, including data privacy issues, lack of compliance with UK/EU regulations, and problematic licensing terms. Institutions must educate researchers about these risks.
Data Privacy Risks: The group highlighted the risks of using free tools that process data outside the EU or do not meet UK data protection standards. Concerns were raised about researchers inadvertently exposing sensitive data or intellectual property.
Licensing Review Toolkit: The University of Birmingham has developed a licensing review toolkit to help academics evaluate the terms and conditions of AI tools before use. This toolkit is now part of their AI training sessions.
Researcher Motivation for Safe Practices: While legal risks may not resonate with researchers, the risk of being ‘scooped’ by competitors is a more effective motivator for encouraging safe practices when using AI tools.
Training Gaps and Behavioural Challenges: While training is essential, it is not sufficient to address all challenges. Many academics may disregard training or fail to engage with institutional resources. A diverse approach involving training, policies, and behavioural change is needed.
Policy Development for AI Use in Research: Several institutions are in the process of developing policies for AI use in research. There is a need for robust policies that address licensing, privacy, data protection, and ethical considerations.
National Collaboration on Licensing: There were discussions about the need for national-level collaboration or bargaining to address the high costs of enterprise-level AI tools like CoPilot and Adobe’s AI solutions.
Research Integrity Policies as a Framework: UKRIO’s guidance on research integrity was highlighted as a potential route for developing generative AI policies. Existing research integrity policies may already implicitly cover aspects of responsible AI use.
Behavioural Change Among Researchers: The need for behavioural change among researchers was emphasised, particularly regarding safe use of AI tools and compliance with institutional policies. Research supervisors may play a key role in influencing these behaviours.
Call for Sharing Resources: Participants expressed interest in sharing resources such as data protection impact assessments (DPIAs) and existing institutional policies on AI use in research to facilitate collaboration and reduce duplication of effort.
Future Developments in AI Tools: Microsoft CoPilot and other platforms are evolving to include more research-specific functionalities. There is potential for broader conversations with tech companies to develop comprehensive research-focused platforms.
We look forward to continuing the conversation at our next meet up which will be taking place on Wednesday December 10th 15:30 – 16:30. We welcome all to join the conversation in shaping what responsible AI in research truly looks like.
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