12/03/2024

Maintaining research integrity in an AI world

Written by Mary Davis

Image credit: UKRIO (2019) The Concordat to Support Research Integrity

I’m looking forward to delivering the next ICAI webinar on Friday December 13 on research integrity and ethical use of AI. The webinar is designed to help researchers, and teaching staff supporting students with research, in maintaining research integrity with the use of AI. I’m writing this as a taster blog for the webinar, to encourage readers to reflect on the core elements of research integrity, the implications of AI use and the need for ethical decision making. I write as a researcher and doctoral supervisor, especially thinking about how I guide students in their use of AI during the research process.
Let’s start by thinking of what research integrity involves. According to the UK Research Integrity Office [UKRIO] (2024) ‘research integrity refers to all of the factors that underpin good research practice and promote trust and confidence in the research process’. This explanation provides a good foundation to understanding that research integrity needs to be fully embedded in every stage of research from planning to dissemination; it is not enough just to think about research integrity in the short term, such as when applying for ethical approval to conduct a study.

UKRIO (2019) define the five core elements of research integrity as: honesty, transparency, accountability, respect and rigour. Drawing on the representation of research integrity in the image above (UKRIO, 2019), we can unpack these elements as follows: Honesty in research means being truthful about how research is conducted, collected and reported. Transparency is about the display of honesty, ensuring that all details are clear to any stakeholder in the research, including declaring conflicts of interest. Accountability moves to thinking about the responsibility of everyone in the research process from the researcher to publisher. Respect involves consideration for everyone and everything in research, especially participants and the environment. Rigour in research, as the final element, is about the care and attention to detail in the methods, analysis and dissemination. These elements of research integrity are of course deeply connected with the Fundamental Values of academic integrity: honesty, trust, fairness, respect, responsibility and courage (ICAI, 2021). With research integrity, it is important to note the additional crucial focus on accountability for those involved and rigour in the research process.

Based on these elements of research integrity, what are the things we need to think about when using AI in research?
For honesty and transparency, all use of AI in research needs to be declared and communicated. It is recommended for student researchers to start by discussing AI use with a supervisor, checking institutional guidance and ensuring declarations are made, perhaps at different stages of research. For all researchers, honesty and transparency about use of AI must be prioritised, especially when findings are shared or disseminated. Furthermore, accountability for use of AI in research involves everyone, including researchers, assistants, institutions, funders and publishers taking responsibility and communicating their use of AI to the others involved. Turning to respect, researchers must have participants’ informed consent to proceed with use of AI, including the uploading of their data to AI tools for analysis or dissemination, which could raise data privacy issues. One growing concern that also relates to respect in research integrity is the impact of AI use on the environment, with the gigantic power utilisation involved; thus, AI use for research should be measured and responsible to limit the environmental impact. Finally, rigour needs to be carefully considered; when involving AI in the research process for data gathering, interpretation and dissemination, the human researcher may struggle to maintain rigour, so building in thorough checking is essential.

How do we apply these elements of research integrity in practice when we use AI? Come along to the webinar for a discussion of ethical scenarios where AI is used in research! Here is the link to register for the ICAI webinar on ‘Research integrity and ethical use of AI’ on December 13, 2024 at 12pm (EST)/ 5pm (UK). I look forward to continuing this discussion with you!

References
International Center for Academic Integrity [ICAI]. (2021). The Fundamental Values of Academic Integrity. (3rd ed.). www.academicintegrity.org/the-fundamental-valuesof-academic-integrity
UK Research Integrity Office [UKRIO] (2024). What is research integrity? Available at https://ukrio.org/research-integrity/what-is-research-integrity/
UK Research Integrity Office [UKRIO] (2019). The UK Concordat to Support Research Integrity. Available at https://ukrio.org/research-integrity/what-is-research-integrity/

The author’s views are their own.

Thank you for being a member of ICAI. Not a member of ICAI yet? Check out the benefits of membership and consider joining us by visiting our membership page. Be part of something great!