Course design maps to stated learning outcomes and the relevant standards, so alignment evidence is produced as a by-product of teaching rather than assembled before an accreditation visit.
The graduate quality question
Ghana’s Minister for Education has publicly raised concerns from the private sector about graduate quality, and urged the Ghana Tertiary Education Commission toward stronger regulatory attention on institutions falling short. That conversation predates generative AI. AI has made it sharper.
Take-home assignments, term papers and much project work no longer evidence the student. A mark awarded for a written submission now certifies what a model produced, filtered through a student’s judgement about what to submit.
Detection is not the answer. False positive rates make detection tools unsafe as the basis of a misconduct finding, and the cost of a wrong accusation is severe. The durable response is changing what is assessed.
Three studios
1. HigherED Pedagogy with AI — design
AI drafts lecture plans, slide decks and course materials mapped to stated learning outcomes and Bloom’s levels. The lecturer edits, owns and signs off.
- Course and session design across 15+ research-driven pedagogies.
- Editable, lecture-ready decks.
- Constructive alignment by construction rather than retrofitted before accreditation.
- Workable for the large lecture cohorts Ghanaian departments actually teach.
2. Spot Observation — develop
Short, evidence-anchored teaching observations scored against a transparent rubric. Every rating requires proof from the lecture room, hedged language is stripped out, and the record includes the lecturer’s right of reply.
The right of reply matters more than the feature list suggests. Teaching observation touches promotion and departmental politics; a record a lecturer cannot answer will not be trusted, and an untrusted observation system produces performances rather than teaching.
3. Innovation Challenge Studio — assess
Students take on authentic, team-based challenges demanding original contribution and external validation, producing portfolio-grade evidence of capability.
| Conventional assessment | Innovation Challenge |
|---|---|
| Recall examinations rewarding memorisation | Authentic briefs requiring original contribution |
| Assignments a model can write in seconds | External validation the student defends in person |
| A single high-stakes mark, no visible process | Process evidenced across the full experiential cycle |
| Certifies what was produced | Evidences what the student can do |
Every challenge ships with a four-band rubric, scaffolded milestones and CV bullet specifications, so assessment measures what graduates can do and employers can trust the record.
Who it is for
- Universities reviewing assessment strategy in response to generative AI.
- Technical universities and polytechnic-heritage institutions with strong practical components.
- Colleges of education, where graduates are themselves future teachers and the observation engine has a double use — particularly relevant given the integration of licensure assessment into final trainee examinations.
- Nursing, engineering and other licensure-track programmes where employer and regulator confidence in the credential is direct.
- Teaching and learning centres and academic planning units.
Accreditation evidence
Running course design, teaching observation and assessment on one record means a stated learning outcome traces from the design that delivered it, through the teaching observed against it, to the student evidence produced. That trace is what accreditation panels ask for and what most departments assemble in a scramble beforehand.
Getting started
Most institutions begin with one department rather than an institution-wide rollout, running the Innovation Challenge Studio on a single course for one cohort with the observation engine alongside. That produces a comparison against the previous cohort within one academic year, which is the evidence a committee will actually act on.