How is AI changing the way organisations in Aotearoa operate, compete, and lead?
Our panel shared practical insights and real-world examples from across Aotearoa, covering everything from Copilot rollouts to predictive modelling. Attendees came away with fresh ideas and a clearer view of what good adoption looked like, how to build internal capability, and what clients were asking for.
Speakers
Maria Mingallon
Maria Mingallon is Mott MacDonald’s AI Lead and Knowledge & Information Manager for the Asia Pacific region. She serves as Deputy Chair of the AI Forum NZ and chairs its AEC Working Group, helping shape the future of AI adoption across the built environment.
Justin Flitter
Justin Flitter founded NewZealand.AI in 2017 to help businesses discover why, where, and how to leverage AI. He is one of NZ’s leading AI-for-business advisors, speakers, and trainers.
Originally published on Building Institute Aotearoa’s website
AI in Action: What Boards and Executives Need to Know (Webinar Summary)
In this Building Institute “Industry Insights” session, AI adviser Justin Flitter (NZ.AI) and Maria Mingalan (Mott MacDonald, AI Forum NZ) break down why the built-environment sector is suddenly moving from “AI curiosity” to “AI capability building” — and what leadership must do to keep it safe, useful, and measurable.
The core driver is blunt: construction businesses have been squeezed, and AI is being treated as a capacity multiplier. The goal isn’t novelty; it’s to win back hours per employee, shift effort away from repetitive admin, and reallocate time into higher-value work that actually moves projects forward. They also warn that adoption may look exponential, not linear — meaning “wait and see” becomes “fall behind and stay behind.”
A recurring theme is that adoption ≠ access. Giving staff a tool (even Copilot) doesn’t mean you’ve adopted AI. Adoption means value realization: measurable time saved, higher quality outputs, faster delivery, and improved staff experience. Maria describes setting explicit KPIs and socializing those metrics with leadership so AI use becomes a visible business program, not a side hobby.
They emphasize responsible AI to prevent “shadow AI”: clarify approved tools, train people on risks and myths (especially data/privacy fears), and build trust with employees and stakeholders. The analogy lands: learning to trust AI outputs is like learning to trust calculators and email when they first arrived — through understanding, not blind faith.
Practical use cases make it concrete: Justin shares a WhatsApp agent for contractors to retrieve job codes and SOPs by voice/text (any language), cutting monthly back-office reconciliation from ~50 hours to ~10. He also highlights meeting transcription + summaries as a way to turn “tribal knowledge” into searchable organizational memory. Maria shares Mott MacDonald examples: chatting with carbon-footprint project data to explore reduction options, and sentiment analysis over large volumes of stakeholder feedback to ensure important signals aren’t missed.
For leaders hiring now, they suggest weaving AI into recruitment: ask candidates how they’ve used AI to reduce task time, automate follow-up work, and critically evaluate outputs — because “accepting the first answer” is the new red flag.
One extra board lens: strong boards focus on the big questions and avoid getting lost in minutiae, which maps directly onto AI governance and ROI priorities.