AI and your Business: Where it Sits, Where it’s Heading, and Why you Can’t Afford to Wait.
Kerstin Littin, Product & Technology
AI has moved from buzzword to boardroom topic in not much more than two years.
Most clients I speak with aren't asking whether AI matters anymore - they're asking how it actually works, where it's heading, and what their business should be doing right now.
Here's how I see it.
A quick word on where this is coming from
Before we get into it, there’s no shortage of people in the New Zealand market right now offering opinions about AI. Most of them are reselling someone else’s tools.
I’ve spent the last twenty years building enterprise software products. As Director of Product at Foster Moore, I led an 18-person team spanning product, UX, marketing, and documentation - building a platform that underpinned $69 million in annual revenue from licensing and government implementation.
The work shipped to some of the most demanding customers in the world: New Zealand’s Ministry of Business, Innovation and Employment; the Australian Taxation Office (over 1 million users); Service Ontario in Canada; the Ministry of Commerce in Cambodia; financial regulators in Botswana and Namibia. The products spanned corporate and occupational registries, KYC, anti-money laundering compliance, and beneficial ownership registers – high stakes systems where "near enough" isn't acceptable.
What that means in practice: I sit at the intersection of product and technology.
I understand both how the tech actually works and how to make it deliver real customer outcomes. That combination is rarer than it should be, and it’s the lens I bring to AI. I’m not selling you a subscription. I’m telling you what I’d do if it were my business.
How AI sits relative to human intelligence today
The AI we all use day-to-day (ChatGPT, Microsoft Copilot, Claude, Google Gemini) is what researchers call narrow AI. It’s extraordinarily capable in specific domains: language, pattern recognition, summarising, coding, image generation, data analysis. In those areas it can already outperform most humans.
But it’s not generally intelligent the way you and I are. It doesn’t truly understand the world, hold beliefs, set its own goals, or reason reliably across unfamiliar situations. Hand it a clear task and it shines. Drop it into a messy, multi-step business problem with shifting goalposts (the kind of work most senior staff do every day) and it still needs human judgement around it.
Think of today’s AI as a brilliant graduate with an exceptional memory and no life experience. Powerful, fast, useful but not autonomous.
The next threshold: AGI
The term you’ll hear a lot more in the next 12 months is AGI (Artificial General Intelligence). This is the point at which AI matches or exceeds human cognitive ability across essentially any task a person can do, not just narrow ones. Beyond AGI sit ASI (Artificial Superintelligence) systems that exceed human intelligence broadly across every domain.
So when does AGI arrive in our day-to-day? Honestly, nobody knows. But the shift in expert opinion is the real story. As recently as 2020, the median forecast was around 50 years away. Today, senior figures at OpenAI, Anthropic, Google DeepMind and Microsoft AI are pointing to somewhere between 2027 and 2033 as plausible. Aggregated forecasts now put roughly a 50% chance of AGI by the early 2030s. Some experts think sooner; some think current approaches will never quite get there. The honest answer is that even the people building it are uncertain, but almost nobody serious is saying “decades away” anymore.
For business, the practical question isn’t whether AGI lands in 2028 or 2035. It’s that AI capability is improving fast enough that planning your operations as if today’s AI is the ceiling is a losing strategy.
This isn't a tech industry story - it's already your industry
If you’re hesitant about AI because it feels like hype, or someone else’s problem, or “not for a business like ours” look at who’s actually using it.
Right here, in New Zealand, today:
Fonterra, Spark, One NZ and Health NZ are rolling out AI capability across their workforces
ASB uses an AI virtual assistant for customer enquiries; Toyota Finance NZ runs intelligent automation across its operations
Foodstuffs has cut food waste by up to 90% in some stores using AI-driven inventory and stock rotation
Vulcan Steel uses Microsoft AI for workplace health and safety monitoring
Vector uses Google’s AI platform alongside drones to inspect Auckland’s electricity network
Halter uses AI-driven collars to manage dairy herds; Robotics Plus runs autonomous AI vehicles in orchards
Xero has built AI directly into its accounting platform, automating coding and reconciliation for thousands of Kiwi businesses
Datacom has rolled AI out to all 6,000 of its staff
And globally, in manufacturing specifically:
Siemens uses AI for predictive maintenance, forecasting equipment failures before they cause downtime
BMW has partnered with Figure AI to bring humanoid robots into mass-production environments
IBM uses generative AI to drive training, quality assurance and regulatory compliance for its manufacturing clients
These aren’t pilots. They’re operating models. And they span every type of business – tier one manufacturers, local distributors, banks, supermarkets, dairy farms, electricity networks.
If you run a business of any kind in New Zealand and you’re still wondering whether AI is “for you,” the honest answer is: your competitors and your suppliers have already decided.
Where New Zealand sits
The wider numbers tell the same story:
Around 87% of New Zealand organisations are now using AI in some form
Of those, 91% report efficiency gains and 77% report reduced operating costs (AI Forum NZ)
91% of Kiwi workers use generative AI to some degree, with more than half using it regularly (Robert Half NZ)
51% of senior business leaders rank AI adoption as their biggest technology opportunity for 2026 (Datacom)
MBIE estimates generative AI could add $76 billion to the New Zealand economy by 2038, that’s about 15% of GDP
The Government has launched an AI Advisory Pilot offering up to $15,000 in co-funding for SMEs to bring in expert help, and is investing $70 million over seven years through the New Zealand Institute for Advanced Technology. Microsoft has committed to AI-skilling 300,000 New Zealanders by 2028.
The pressure point: only around a quarter of New Zealand workers have had any formal AI training. We have the access; we don't yet have the capability - and that gap is the opportunity for businesses willing to move now.
The next 24 months will separate the leaders from the laggards
Here's the part I want to be direct about.
The next two years are going to sort New Zealand businesses into those that come out of this stronger and those that fall behind permanently. This happens with every major technology shift. The internet did it in the 90s. Cloud did it in the 2010s - I lived through that one, leading the migration of multiple government registries from legacy on-premise systems onto cloud-native platforms. The pattern is identical, and the reason is always the same: the businesses that wait don’t get a discount when they finally move; they just inherit a bigger gap to close.
AI is doing the same thing now and the timeline is shorter than either of the previous waves, because the technology is improving in months rather than years.
If you run a manufacturing business, a trades business, a distribution business, a professional services firm - the temptation is to treat AI as a future problem, or to wait for it to “settle down.” It isn’t a future problem. It’s a 2026 problem. Within 24 months, your competitors will be quoting faster, responding to customers faster, running tighter inventory, producing better proposals, and operating with lower overheads than you are - using tools that are available to you right now, often for under $5,000 to set up.
The risk isn’t moving too fast on AI. The risk is letting two years go by while everyone else builds capability you don’t have.
What to do, practically
Stop treating AI as an IT project. It’s a productivity and operating-model question. The biggest wins are in finance, marketing, customer service, admin and operations - not your tech team.
Pick one workflow and prove value. One real workflow - quoting, reporting, customer responses, meeting notes, reconciliations - and a measurable saving. Microsoft’s modelling suggests workers can claw back around 275 hours per person per year. Capture even part of that and you’ve funded the next initiative.
Invest in capability, not just tools. The gap between businesses that benefit and those that don’t is narrowing on technology and widening on training.
Use what’s already on the table. The AI Advisory Pilot, the NZ Institute for Advanced Technology, and the major skilling programmes from Microsoft and others are real levers most SMEs aren’t yet pulling.
You don't need to become a technologist. You need to engage.
How we can help - AI Ready
This is the work we do at Blomfield. Our AI Ready programme takes New Zealand businesses from “AI curious” to “AI capable” - practically, hands-on, and with no vendor relationships influencing the advice you get. That last point matters. Most “AI consultants” are recommending the tools that pay them a referral fee. We don’t.
The programme is built on product management discipline - discovery, iteration, and measuring real outcomes against real objectives. That’s the same approach used to ship the world’s most successful software products, and it’s what separates AI work that delivers from AI work that produces a slide deck.
It runs in three stages, each at a fixed fee so you know exactly what you’re committing to:
1. Assessment
An honest, independent view of where AI can genuinely move the dial in your business
2. Roadmap
A written 90-day and 12-month implementation plan, presented to your leadership team
3. Implementation
We roll up our sleeves and work alongside your team to put it in place (three packages: Kickstart, Accelerator, Transformation)
Existing Blomfield clients receive a 10% discount on Stage 1.