Bring a brief and someone to review it.
When choosing AI role play software for a university, check eight things: curriculum-level scenarios, scoring you can inspect, faculty oversight, LMS launch, data policy, language coverage, proctoring and adoption support. Run your own scenario with real students before committing.
- An institution-owned task and rubric
- Faculty and technical reviewers
- Your procurement and data requirements
5 moves to practise
Curriculum fit
Are scenarios built for your programme with subject experts, or generic conversations with a new label? Ask to see criteria for your subject.
Scoring and oversight
Can faculty see the rubric, review recordings, override scores and handle disputes? Marks without evidence will not survive an appeal.
Integration and access
Does it launch through LTI 1.3 in your LMS, and can a student start from a link? Extra logins kill adoption.
Data and security
Is student data used to train models or sold? Ask for a clear answer and put it in the agreement.
Adoption
Student use, not licence spend, is the usual worry. Ask what handholding is included and how usage is shown to faculty.
A procurement committee discusses a trial
Illustrative scriptHow will we know this works?
We pick one scenario and sixty students, and review attempts against our own rubric.
What decides yes or no?
Student use, faculty review of a sample and the evidence we can show.
Quick scorecard
| Area | Ask the vendor |
|---|---|
| Curriculum | Show criteria for our subject |
| Evidence | Can we see a recording and rubric for one student? |
| LMS | Is it LTI 1.3 and have you tested our LMS? |
| Data | Is student data used for training or shared? |
| Languages | Which languages and accents are supported? |
| Adoption | What support do we get in the first month? |
Try it. Review it. Change one thing.
Your practice task
Ask each provider to demonstrate the same task and explain what evidence reviewers can inspect.
What to look for
Record what was demonstrated, what remains unverified and who owns each follow-up. Agree decision criteria before comparing offers.
Common questions
Should we run a pilot?
Test one real scenario with a small group of students and compare feedback with faculty.
Is price the main factor?
Adoption and evidence matter more than per-session price.
Does Teachrity meet this checklist?
See how Teachrity works and ask us to demonstrate any item.
Research and further reading
Use these primary resources to explore the method and its context. References describe their publishers’ work; they do not imply endorsement or certification of Teachrity.
- Cornell: designing oral assessments
Practical assessment-design considerations from Cornell’s Center for Teaching Innovation.
- AACSB: assurance of learning
Evidence and improvement within AACSB’s own accreditation framework; using software does not confer accreditation.