Enterprise AI adoption in New Zealand and Australia is at an inflection point. The early hype has given way to pragmatism, and the gap between organisations experimenting with AI and organisations deploying it at scale is widening. This report synthesises publicly available data, international benchmarks, and our direct experience across the NZ/AU enterprise market.
Executive Summary
The NZ/AU enterprise AI picture in late 2024 can be characterised by three themes:
- High awareness, low maturity. Most enterprises acknowledge AI's importance but few have moved beyond pilots.
- Talent bottleneck. The skills gap is the primary constraint, not budget, not technology, not executive buy-in.
- Foundation gap. Organisations deploying AI successfully have invested in shared infrastructure. Most haven't.
89%
of NZ/AU enterprises say AI is strategically important
Source: Compiled from NZTech and CSIRO reports, 2024
23%
have deployed AI in production beyond a single use case
Source: Compiled from NZTech and CSIRO reports, 2024
The gap between "strategically important" (89%) and "actually deployed at scale" (23%) is the defining metric of the current moment.
The Enterprise AI Intention-Action Gap
Source: Compiled from NZTech and CSIRO reports, 2024
Adoption by Sector
Financial Services - Leading
Banks and insurance companies are the furthest ahead, driven by clear ROI use cases (fraud detection, claims processing, credit risk) and regulatory pressure to adopt responsibly.
NZ's major banks have active AI programmes, though most are still concentrated in a small number of use cases rather than deployed organisation-wide.
Government - Cautious but Moving
NZ government agencies are proceeding carefully, constrained by the Algorithm Charter, public accountability, and Te Tiriti obligations. Several agencies have significant AI pilots in areas like biosecurity, health, and social services.
The pace is deliberately slower than the private sector, but the quality of governance frameworks is often higher.
Healthcare - Early Stage
Despite massive potential, NZ/AU healthcare AI adoption remains early. Regulatory caution, data fragmentation, and workforce concerns are the primary barriers. The exceptions are radiology AI (where regulatory pathways exist) and administrative AI (scheduling, coding, documentation).
Professional Services - Experimental
Law firms, accounting practices, and consultancies are actively experimenting with generative AI for document drafting, research, and analysis. Most adoption is at the individual or team level rather than firm-wide.
56%
of NZ professional services firms report staff using AI tools, but only 12% have firm-wide policies
Source: NZLS Technology Survey, 2024
Manufacturing and Primary Industries - Nascent
AI adoption in NZ manufacturing and primary industries (agriculture, forestry, fisheries) is nascent but promising. Use cases focus on predictive maintenance, quality control, and supply chain optimisation.
Australia is further ahead here, driven by mining and resources sector investment.
The Talent Bottleneck
68%
of NZ enterprises cite talent as their primary AI constraint
Source: NZTech, AI Skills Gap Report 2024
The NZ/AU talent market for AI is acutely constrained. The skills gap operates at three levels:
Technical Talent
ML engineers, data scientists, and AI architects are in global demand. NZ and Australian salaries compete with Singapore and parts of Asia but not with US tech company compensation. Many trained AI professionals leave for offshore opportunities.
Applied Talent
The bigger gap is in people who can bridge AI capability and business problems. Product managers who understand AI limitations. Business analysts who can specify AI use cases. Change managers who can help teams adopt AI tools.
Leadership Talent
Executives who understand AI deeply enough to govern it, invest in it wisely, and hold vendors accountable. This is perhaps the scarcest capability of all.
The talent conversation focuses too much on data scientists and too little on the leadership and applied skills that determine whether AI investments deliver value. You need 5 leaders who understand what AI can and can't do.
Dr Tania Wolfgramm
Chief Research Officer
What Leading Organisations Do Differently
Across our experience and the available research, the organisations successfully deploying enterprise AI share five characteristics:
1. They Invest in Foundations, Not Just Projects
Leading organisations build shared AI infrastructure (data pipelines, model hosting, governance frameworks) that every team can use. Each new AI project builds on what came before.
2. They Start with Business Problems, Not Technology
The failed pattern: "We have GPT-4, what should we do with it?" The successful pattern: "We lose $2M annually to manual claims processing. Can AI help?"
3. They Govern Proactively
They don't wait for regulation. They build governance frameworks early, use them as competitive differentiators, and adapt as the regulatory environment evolves.
4. They Build Internal Capability
They don't outsource AI entirely. They use partners to accelerate, but they build internal teams who understand, maintain, and extend AI systems.
5. They Measure Ruthlessly
They define success metrics before starting, measure throughout, and kill projects that don't deliver. They don't let sunk cost or executive enthusiasm keep failing projects alive.
NZ/AU vs Global
How does NZ/AU compare to global enterprise AI adoption?
| Dimension | NZ/AU | Global Average | Leading Markets (US, UK, Singapore) |
|---|---|---|---|
| Executive awareness | High (89%) | High (85%) | Very High (94%) |
| Production deployment | Low (23%) | Moderate (35%) | High (52%) |
| Governance frameworks | Low (34%) | Moderate (38%) | Moderate-High (47%) |
| Talent availability | Constrained | Constrained | Less constrained |
| Investment per enterprise | Lower | Moderate | Higher |
NZ/AU is roughly 12-18 months behind leading markets in production deployment, but the gap is closing. The smaller market size means individual organisations can move faster when they commit.
Production AI Deployment by Market
Source: Compiled from NZTech, CSIRO, McKinsey, and Gartner reports, 2024
The NZ/AU Advantage
Smaller market, faster decisions. NZ/AU enterprises can implement AI at a pace that would require 12 months of stakeholder management in a Fortune 500. The organisations that recognise this speed advantage, and pair it with governance maturity, will punch above their weight globally.
Outlook for 2025
Three trends will shape NZ/AU enterprise AI in 2025:
1. Generative AI moves from experimentation to integration. The novelty phase is over. 2025 is when generative AI either becomes part of business workflows or gets labelled as overhyped.
2. Governance becomes competitive advantage. As regulation approaches (driven by EU AI Act precedent and Australian legislative development), organisations with governance frameworks already in place will move faster than those scrambling to comply.
3. The foundation builders pull ahead. The gap between organisations with shared AI infrastructure and those running isolated pilots will widen significantly. The compound advantage becomes visible.
2024 was the year NZ enterprises decided AI matters. 2025 is the year they'll find out whether they've built the foundation to deliver on that decision.
Isaac Rolfe
Managing Director
- Is NZ too small for enterprise AI to be worthwhile?
- No. In fact, NZ's size is an advantage. Smaller organisations make faster decisions, have shorter feedback loops, and can iterate more quickly. The challenge isn't market size. It's talent availability and the tendency to wait for larger markets to prove concepts before adopting.
- What's the average enterprise AI investment in NZ?
- Direct data is limited, but based on our experience, NZ enterprises typically invest $200K-$2M in their first significant AI initiative. This is lower than US/UK averages but delivers comparable results when focused on high-value use cases.
- Which NZ sectors will adopt AI fastest?
- Financial services will continue to lead, followed by government (where AI can improve service delivery at scale). Healthcare has the highest long-term potential but faces the most regulatory and data barriers. Professional services adoption will accelerate as generative AI tools mature.