Decisions to (dis)engage with AI are being made all around us, by policy makers, practitioners in a wide range of fields including education, and of course by students and people in their everyday lives. I say (dis)engage to emphasise that when we talk about AI, and AI ethics, that discussion should reflect that sometimes the ethical decision is to not engage or to disengage, rather than trying to work out the most ethical way to engage (where that may not be possible).
To make such decisions, learning is required:
- Decision makers have to learn about AI systems and their suitability to address the needs they’re being set to,
- AI developers need to learn how to engage with and communicate this as well as how to provide such information to end-users,
- end-users or others impacted by AI use have to learn about what AI is being used and why, how to use (or not use) these tools, and how to become involved in these processes.
This is hard, it involves learning about different areas across stakeholders who are not experts in (variously) areas of policy or practice, AI development and evaluation at its various stages, and areas of ethics or understanding of local impacts of tools. In a set of recent publications and policy contributions, I have been trying to set out some of my thinking on this issue, and framing the centrality of learning in thinking about ethical (dis)engagement with AI. This learning matters. Those in positions of power have a duty to learn about the issues worth, and not worth, directing AI towards, and this requires learning with stakeholders to consider impacts of decisions and tools. Moreover, ethical (dis)engagement with AI necessarily involves support for learning: through engagement in co-design to identify issues worth addressing and approaches to address them; in simple acts such as providing informed consent or actually understanding explainable (explained?) AI; and more broadly in the ways that acting as an informed citizen requires understanding of AI.
Policy impact
A set of recent policy contributions (to which I have made greater and lesser contributions) have spoken to some of these issues, including those below.
- Gill, A., Heggart, K., Johns, A., Knight, S., Narayan, B., & Shibani, A. (2024). UNSW-UTS:Trustworthy Digital Society submission in response to the House Standing Committee on Employment, Education and Training’s inquiry into the Digital Transformation of Workplaces (Report No. 12). House Standing Committee on Employment, Education and Training. https://opus.lib.uts.edu.au/handle/10453/179650
- Gulson, K., Swist, T., Beehag, L., Villarino, J.-M. B., Thompson, G., Zhang, V., Knight, S., & Kitto, K. (2023). Education Futures Studio and Sydney Policy Lab Submission in response to the House Standing Committee on Employment, Education and Training’s inquiry into the use of generative artificial intelligence in the Australian education system [Report]. House Standing Committee on Employment, Education and Training. https://education-futures-studio.sydney.edu.au/learning/#white-papers-reports
- Heggart, K., Knight, S., Maher, D., & Narayan, B. (2024). UTS:CREDS Submission in response to the Joint Select Committee on Social Media and Australian Society (Report No. 83). Joint Select Committee on Social Media and Australian Society. https://opus.lib.uts.edu.au/handle/10453/179772
- Knight, S. (2024). Submission in response to the Joint Standing Committee on Electoral Matters’ Inquiry into civics education, engagement, and participation in Australia (Report No. Sub 23). Joint Standing Committee on Electoral Matters. https://opus.lib.uts.edu.au/handle/10453/179100
- Knight, S., Heggart, K., Dickson-Deane, C., Ford, H., Hunter, J., Johns, A., Kitto, K., Cetindamar Kozanoglu, D., Maher, D., & Narayan, B. (2023). UTS:CREDS Submission in response to the House Standing Committee on Employment, Education and Training’s inquiry into the use of generative artificial intelligence in the Australian education system (Report No. 21). House Standing Committee on Employment, Education and Training. https://opus.lib.uts.edu.au/handle/10453/171583
- NTEU. (2024). Discussion Paper on the Impacts of Artificial Intelligence Technologies in the Tertiary Education Sector (No. Contributor to). National Tertiary Education Union. https://nteuau.sharepoint.com/teams/NationalOffice/Administration/Forms/AllItems.aspx?id=%2Fteams%2FNationalOffice%2FAdministration%2FNCM%2FNCM%202024%20Online%2F3%2E%20Motions%20on%20Notice%2FD2%20Higher%20Education%20Policy%20and%20Research%2FNTEU%20AI%20Discussion%20Paper%202024%2Epdf&parent=%2Fteams%2FNationalOffice%2FAdministration%2FNCM%2FNCM%202024%20Online%2F3%2E%20Motions%20on%20Notice%2FD2%20Higher%20Education%20Policy%20and%20Research&p=true&ga=1
- UTS. (2023). UTS Submission in response to the House Standing Committee on Employment, Education and Training’s inquiry into the use of generative artificial intelligence in the Australian education system [Report]. House Standing Committee on Employment, Education and Training. https://opus.lib.uts.edu.au/handle/10453/175054
It’s hard to track the impact or influence of these things, but I am pleased that they do seem to have reached some audience, with the work being drawn on for example in the below.
- House Standing Committee on Employment, Education and Training. (2024). Study Buddy or Influencer: Report of The Inquiry into the Use of Generative Artificial Intelligence in the Australian Education System (Australia; Inquiry into the Use of Generative Artificial Intelligence in the Australian Education System) [Text]. Australian Government. https://www.aph.gov.au/Parliamentary_Business/Committees/House/Employment_Education_and_Training/AIineducation/Report
- McIntyre, F. (2024, September 20). Protecting science: Why academics want support on AI. Research Professional. https://www.researchprofessional.com/0/rr/news/australia/politics/2024/9/Protecting-science—why-academics-want-support-on-AI.html
- NHMRC. (2024, May 29). NHMRC Artificial Intelligence Workshop Report. Australian Government. https://www.nhmrc.gov.au/about-us/publications/nhmrc-artificial-intelligence-workshop-report
Research underpinnings
In a set of recent pieces I (with my fantastic colleagues) have explored some of the issues reflected in those policy contributions.
Each of these is detailed below with the papers addressing:
- Papers on the material resources for AI ethics and governance, analysing: a. the research ethics committee processes; b. editorial policies and how they can foster learning regarding ethics; c. how ethics guidelines are developed and the role of evidence (and evidence synthesis) in this, particularly with respect to AI in education; d. how AI is represented in submissions to a national public inquiry on AI in education.
- Consideration of the role of material cases to support learning particularly regarding tradeoffs in AI ethics, and the materials (e.g., data) of AI and the implications of this for AI uses.
1. Papers on the material resources for AI ethics and governance, analysing:
a. the research ethics committee processes; b. editorial policies and how they can foster learning regarding ethics; c. how ethics guidelines are developed and the role of evidence (and evidence synthesis) in this, particularly with respect to AI in education; d. how AI is represented in submissions to a national public inquiry on AI in education.
- Knight, S., Shibani, A., & Vincent, N. (2024). Ethical AI governance: Mapping a Research Ecosystem. AI and Ethics. https://doi.org/10.1007/s43681-023-00416-z
- Knight, S., Viberg, O., Mavrikis, M., Kovanović, V., Khosravi, H., Ferguson, R., Corrin, L., Thompson, K., Major, L., Lodge, J., Hennessy, S., & Cukurova, M. (2024). Emerging technologies and research ethics: Developing editorial policy using a scoping review and reference panel. PLOS ONE. https://doi.org/10.1371/journal.pone.0309715
- Knight, S. (2024). Is that a Guideline? A PRISMA-ETHICS informed Scoping Review of Ethics Guideline Development Approaches, Structures, and Audiences (No. n43d6). OSF Preprints. https://doi.org/10.31219/osf.io/n43d6
- Knight, S., Dickson-Deane, C., Heggart, K., Kitto, K., Çetindamar Kozanoğlu, D., Maher, D., Narayan, B., & Zarrabi, F. (2023). Generative AI in the Australian Education System: An Open Dataset of Stakeholder Recommendations and Emerging Analysis from a Public Inquiry. Australasian Journal of Educational Technology (AJET), 39(5), 101–124. https://doi.org/10.14742/ajet.8922
2. Consideration of the role of material cases to support learning particularly regarding tradeoffs in AI ethics, and the materials (e.g., data) of AI and the implications of this for AI uses.
- Knight, S., McGrath, C., Viberg, O., & Pargman, T. C. (2025). Learning about AI ethics from cases: A scoping review of AI incident repositories and cases. AI and Ethics. https://doi.org/10.21203/rs.3.rs-4844649/v1
- Swist, T., Gulson, K., Knight, S., Benn, C., Kitto, K., O’Callaghan, S., Reid, A., & Zhang, V. (2024). A technical democracy design experiment: Making the UK exam algorithm controversy game. Design Studies, 91–92, 101245. https://doi.org/10.1016/j.destud.2024.101245
- Knight, S., Ford, H., Shibani, A., & Chambers, S. (2023). Inverting {{Citation needed}}: A Critical Design Reflection. Learning, Media and Technology, 49(4). https://doi.org/10.1080/17439884.2023.2244412
- Knight, S., Shibani, A., & Buckingham Shum, S. (2023). A Reflective Design Case of Practical Ethics in Learning Analytics. British Journal of Educational Technology. https://doi.org/10.1111/bjet.13323
- Kitto, K., & Knight, S. (2019). Practical ethics for building learning analytics. British Journal of Educational Technology, 50(6), Article 6. https://doi.org/10.1111/bjet.12868
- Cerratto Pargman, T., McGrath, C., Viberg, O., & Knight, S. (2023). New vistas on responsible learning analytics: A data feminism perspective. Journal of Learning Analytics, 1–16. https://doi.org/10.18608/jla.2023.7781