For the past semester, RIVER has been co-lecturing a postgraduate course at the University of Auckland, teaching AI strategy and application to Māori entrepreneurs through MAORIDEV 721. The experience has challenged nearly every assumption we held about how AI should be taught in a business context, and reshaped how we think about knowledge transfer more broadly.
What You Need to Know
- Teaching AI to entrepreneurs is fundamentally different from teaching it to technologists. The questions are better: "What problem does this solve for my community?" vs "What model should I use?"
- Māori entrepreneurship brings a perspective that mainstream AI education lacks: long-term thinking, community-first design, and a healthy scepticism about technology for its own sake.
- The biggest gap in AI education isn't technical literacy. It's business application literacy. People don't need to understand transformer architecture. They need to understand what AI can and can't do for their specific context.
- Co-lecturing (practitioner + academic) is a powerful model. Theory without practice is abstract. Practice without theory is anecdotal. Together, they stick.
Why We're in the Classroom
RIVER has been delivering AI in enterprise settings for over a year now. We've seen what works, what fails, and what the gap is between AI hype and AI reality. When the opportunity came to co-lecture alongside Dr Carla Houkamau's team at the University of Auckland Business School, we saw a chance to test whether our practical experience could translate into education.
MAORIDEV 721 is a postgraduate course for Māori entrepreneurs, many of whom are building businesses with deep connections to their communities, iwi, and whanau. These aren't abstract case studies. These are people building real ventures where AI might (or might not) play a role.
What Surprised Us
The Questions Were Better
In enterprise settings, the first question is usually "What AI tool should we buy?" In the classroom, the first question was "Why would my customers care about this?" That's a better starting point.
Entrepreneurs think in terms of value delivered and problems solved. They don't have the luxury of technology for its own sake. Every tool has to justify itself against the alternative of not using it. This filters out a lot of AI noise very quickly.
Cultural Context Changes Everything
Mainstream AI discourse is overwhelmingly Western, Silicon Valley-flavoured. It assumes certain things about data availability, individual consent, and the primacy of efficiency. Māori entrepreneurship operates from different foundations: collective benefit, intergenerational thinking, and kaitiakitanga (guardianship).
These aren't constraints on AI adoption. They're design requirements. An AI tool for a Māori health provider doesn't just need to be accurate. It needs to respect whakapapa, handle sensitive cultural knowledge appropriately, and serve the community's long-term interests, not just the organisation's short-term efficiency.
This pushed us to think harder about our own work. How much of our enterprise AI practice assumes a narrow definition of "value"?
Practical Beats Theoretical Every Time
The sessions that worked best weren't the ones where we explained how language models work. They were the ones where students brought their own business problems and we worked through whether and how AI could help.
One student was building a platform for Māori language revitalisation. Another was developing a food business with traditional supply chains. Another was working on environmental monitoring. In each case, the AI conversation was completely different, not because the technology changed, but because the context changed everything about what "useful" means.
26
postgraduate students in the MAORIDEV 721 cohort, representing ventures across health, education, food, and technology
Source: University of Auckland Business School, 2024
What This Teaches Us About Enterprise AI
The classroom insights translate directly to enterprise:
Start with the problem, not the technology. This isn't new advice, but watching entrepreneurs naturally do this, while enterprise teams naturally don't, is instructive.
Context is the strategy. Two businesses in the same industry might need completely different AI approaches based on their values, their customers, and their operating model. "Best practice" is a starting point, not a destination.
AI literacy doesn't mean technical literacy. The most effective AI decision-makers we've worked with (in the classroom and in enterprise) aren't the most technical. They're the ones who can clearly articulate what good looks like and assess whether the AI is delivering it.
Long-term thinking produces better AI strategy. When you're thinking about the next quarter, AI is a tool. When you're thinking about the next generation, AI is infrastructure. The long-term perspective leads to better architectural decisions.
What's Next
We're continuing the co-lecturing relationship into the next academic year. The plan is to develop case studies from this cohort's experiences that can inform both academic curricula and our own enterprise practice.
If there's one thing we've taken away from this experience, it's humility. We walked in thinking we'd teach AI. We walked out having learned at least as much about business, community, and the limitations of a purely commercial lens on technology.
The best AI learning session I have been part of this year was not in a boardroom. If your AI strategy cannot answer "why would my community care?", it probably needs more work.
Dr Josiah Koh
Education & AI Innovation