Appendix D — Aligning the Activity with Institutional Outcomes

Use this appendix to connect an activity to the actual programme, unit and assessment requirements. AI use does not itself demonstrate alignment, effectiveness or accreditation compliance.

D.1 Begin with the local requirements

Locate the unit outcomes, programme-level assurance approach, assessment policy, approved tool/data guidance and access arrangements. Record the relevant version and who can resolve uncertainty.

The TEQSA-commissioned 2025 resource describes several institutional approaches to assurance. Use it to inform a local design discussion, not as a claim that one transcript-based assignment satisfies all requirements.

D.2 A worked mapping

The PDP pilot has a deliberately narrow capability: plan and justify a fair first response using supplied records.

Outcome element	Practice	Evidence to inspect
Use evidence	Annotate P1 records	Claims linked to source IDs
Respect process	Compare P1-A/P1-B	Response opportunity and authority
Evaluate feedback	Check P1-F	Accepted/rejected advice with reasons
Apply independently	Attempt P2	Independent referral decision
Communicate	Explain next step	Clear, bounded wording

The mapping does not establish competence in a whole professional role. If a programme requires live consultation skill, add an appropriate observed interaction. If it requires actual legal knowledge, supply reviewed legal teaching and assess that knowledge rather than the invented practice policy.

D.3 Questions for the teaching team

Which capability is essential? Where else is it taught and assessed? Does this activity add useful practice or duplicate an existing task? What independent evidence will support progression decisions? Can students demonstrate the outcome through equitable routes?

Check that AI-specific criteria appear only where AI use is itself an outcome. A student should not lose disciplinary marks simply because they worked independently and submitted a clear evidence note rather than a chat.

D.4 Approval and workload record

Record the proposed activity, outcome, assessment conditions, criteria and data decisions. Identify the owner of access support, moderation and incident handling. Budget design, testing, support and marking time.

State what is proposed and what has been observed. A small trial can identify access obstacles or persistent errors; it cannot by itself establish general effectiveness across programmes.

Where the activity changes a graded assessment, follow the institution’s approval and student-notification processes. A lecturer’s enthusiasm does not override published assessment conditions.

D.5 Activity: explain the alignment

Allow ten minutes; no AI required. Write a short justification for one activity you plan to use. Name its outcome, the evidence to inspect, the independent assurance point and one limitation.

A useful justification can support either adopting AI or retaining a non-AI route. Its strength comes from the learning design, not the presence of a tool.

The workshop guide provides a way to review that design with colleagues before launch.

# Aligning the Activity with Institutional Outcomes {#sec-alignment} Use this appendix to connect an activity to the actual programme, unit and assessment requirements. AI use does not itself demonstrate alignment, effectiveness or accreditation compliance. ## Begin with the local requirements Locate the unit outcomes, programme-level assurance approach, assessment policy, approved tool/data guidance and access arrangements. Record the relevant version and who can resolve uncertainty. The [TEQSA-commissioned 2025 resource](https://www.teqsa.gov.au/sites/default/files/2025-09/enacting-assessment-reform-in-a-time-of-artificial-intelligence.pdf) describes several institutional approaches to assurance. Use it to inform a local design discussion, not as a claim that one transcript-based assignment satisfies all requirements. ## A worked mapping The PDP pilot has a deliberately narrow capability: plan and justify a fair first response using supplied records. | Outcome element | Practice | Evidence to inspect | |---|---|---| | Use evidence | Annotate P1 records | Claims linked to source IDs | | Respect process | Compare P1-A/P1-B | Response opportunity and authority | | Evaluate feedback | Check P1-F | Accepted/rejected advice with reasons | | Apply independently | Attempt P2 | Independent referral decision | | Communicate | Explain next step | Clear, bounded wording | The mapping does not establish competence in a whole professional role. If a programme requires live consultation skill, add an appropriate observed interaction. If it requires actual legal knowledge, supply reviewed legal teaching and assess that knowledge rather than the invented practice policy. ## Questions for the teaching team Which capability is essential? Where else is it taught and assessed? Does this activity add useful practice or duplicate an existing task? What independent evidence will support progression decisions? Can students demonstrate the outcome through equitable routes? Check that AI-specific criteria appear only where AI use is itself an outcome. A student should not lose disciplinary marks simply because they worked independently and submitted a clear evidence note rather than a chat. ## Approval and workload record Record the proposed activity, outcome, assessment conditions, criteria and data decisions. Identify the owner of access support, moderation and incident handling. Budget design, testing, support and marking time. State what is proposed and what has been observed. A small trial can identify access obstacles or persistent errors; it cannot by itself establish general effectiveness across programmes. Where the activity changes a graded assessment, follow the institution's approval and student-notification processes. A lecturer's enthusiasm does not override published assessment conditions. ## Activity: explain the alignment **Allow ten minutes; no AI required.** Write a short justification for one activity you plan to use. Name its outcome, the evidence to inspect, the independent assurance point and one limitation. A useful justification can support either adopting AI or retaining a non-AI route. Its strength comes from the learning design, not the presence of a tool. The [workshop guide](workshop-guide.qmd) provides a way to review that design with colleagues before launch.