The Hidden Cost Of AI Isn't The Technology
The Atamai Perspective
Anne McGuire · Founder, Atamai Advisory · August 2026
If your organisation is budgeting for AI, there's a good chance you're budgeting for the cheapest part.
At last week's AI Innovation Summit, Professor Jon Whittle referenced Stanford research suggesting that for every $1 invested in AI technology, organisations should expect to invest closer to $10 in the people, process and organisational change required to realise value.
That statistic stayed with me because it completely reframes how we should think about AI investment.
Just this week Canva reported that the rising cost of building products using frontier AI models had contributed to product launch delays and revenue coming in below earlier guidance. For one of Australia's most successful technology companies, the economics of AI are becoming a genuine business consideration.
Taken together, those two observations point to a much bigger shift.
The conversation is no longer just about what AI can do. It's about what it takes to create value from it.
The AI industry has invested more than US$1.4 trillion building the future, yet it has recovered only a fraction of that investment. There's a website called isaiprofitable.com that tracks these numbers in real time, and it offers a fascinating view of where today's AI economy is creating value.
The company generating the strongest returns is NVIDIA, supplying the chips that power almost every major AI model. It's a familiar pattern. During a gold rush, the people selling the shovels often make more money than the prospectors.
It's easy to look at these numbers and conclude that AI is simply expensive. I think that's missing the point. The real cost of AI isn't the licence, the subscription or the API call, it's everything that comes afterwards.
It's redesigning business processes so AI fits naturally into the way work gets done. It's preparing data, establishing governance, building new capabilities, training teams and helping leaders navigate organisational change. And it's measuring outcomes and continuously refining how AI is used across the business.
In other words, AI isn't primarily a technology initiative.
It's an organisational transformation initiative enabled by technology. That also explains why so many organisations struggle to realise meaningful returns from their AI investments.
Most conversations still begin with technology: Which model should we use? Should we buy Copilot? Should we build an AI agent? Which platform is best?
They're all reasonable questions, but they're rarely the first questions organisations should be asking.
The most interesting conversations I had during the Summit weren't about large language models. They were about business problems.
Manual order processing.
Customer service teams overwhelmed by repetitive enquiries.
Knowledge locked away in people's inboxes.
Legacy systems that no longer support the way the business wants to operate.
Hours of administrative work that add little value but consume significant time every week.
In every conversation, AI isn't the objective, improving the business is; that's an important shift.
Organisations are moving beyond simply wanting to "use AI". They're starting to ask where AI genuinely fits within their operating model, which processes should change, where people continue to add the greatest value, and how they'll measure whether the investment is delivering meaningful business outcomes.
At the same time, AI is changing the economics of software itself.
More organisations are questioning whether they need another off-the-shelf application or whether they can build targeted AI solutions to solve specific business problems. AI has dramatically lowered the barrier to creating software, but it hasn't removed the cost of owning it. Whether you're paying for licences, API usage, cloud infrastructure or internal capability, someone is still paying for every workflow, every prompt and every automated decision.
Perhaps that's the biggest lesson from both Canva's announcement and the conversations throughout the AI Innovation Summit.
The AI licence isn't your AI investment, it's simply the entry point. The real investment is everything required to change how your organisation operates.
Before approving another AI initiative, I think leadership teams should be asking questions such as:
What business problem are we solving?
How will this change the way people work?
What investment will be required beyond the technology itself?
How will we measure whether it's delivering value?
If it's successful, how will we scale it across the organisation?
The organisations that create the greatest value from AI won't necessarily be the ones spending the most on technology.
They'll be the ones investing in the capability to use it well because that's where competitive advantage is created.
About Anne McGuire
Anne McGuire is Founder of Atamai Advisory, an executive advisory practice helping organisations navigate growth, transformation and AI through better operating models, organisational capability and practical execution.
Anne writes The Atamai Perspective, sharing observations from more than two decades leading operations, transformation and technology across global enterprises and high-growth businesses.
Work With Atamai Advisory
If your organisation is exploring AI but wants to ensure the investment translates into meaningful business outcomes, let's talk.