AI Capability Checkpoints

This document provides optional reflection scaffolds aligned to a six-domain AI Capability model.
These checkpoints are designed to support responsible, transparent, and professional research practice.

Completion is not required.
If AI tools were not used in your research process, this section may be left blank or removed.

1. Awareness & Orientation

If AI tools were used at this stage of the project:

  • What tools were used (e.g., drafting assistants, coding support, summarisation tools)?
  • What are their known limitations (e.g., hallucination, bias amplification, incomplete context)?
  • What checks were performed to ensure accuracy and reliability?

If no AI tools were used, briefly state that the work was completed without AI assistance.

2. Human–AI Co-Agency

If applicable:

  • Which tasks were supported by AI tools?
  • Which decisions remained fully under human judgement?
  • Where was human oversight explicitly exercised?
  • How were AI-generated suggestions evaluated or rejected?

Clarify that ultimate responsibility for the work remains with the research team.

3. Applied Practice & Innovation

If AI tools were used:

  • In what specific ways did AI support drafting, structuring, analysis, or critique?
  • What elements of the research were developed independently?
  • Were any AI outputs modified substantially before inclusion?

Avoid overstating AI contributions. Emphasise verification and professional judgement.

4. Ethics, Equity & Impact

If AI tools were involved:

  • Were any sensitive or identifiable data shared with AI systems?
  • What bias or fairness risks were considered?
  • Could AI-assisted language or framing unintentionally misrepresent findings?
  • How were equity implications reviewed?

If AI was not used, consider whether the research topic itself carries equity or impact considerations.

5. Decision-Making & Governance

If AI tools were used:

  • How was AI use documented (e.g., in decision logs or methods notes)?
  • Were institutional, funder, or journal disclosure requirements considered?
  • How were AI-assisted decisions justified and recorded?

Ensure transparency and traceability where relevant.

6. Reflection, Learning & Renewal

Optional reflective prompts:

  • What worked well (or not) when integrating AI into this project?
  • What would you do differently in future work?
  • Did AI influence the framing, clarity, or structure of the research?

If AI was not used, reflect briefly on whether it would have added value or risk in this context.

Important Notes

  • AI use is not required for this template.
  • These checkpoints are designed to support transparency and governance.
  • Researchers may adapt, shorten, or remove this section to suit their workflow.
  • Responsibility for all research outputs remains with the human authors.