Two AI Conversations: What Australia and Singapore Are Telling Me About the Next Phase of Transformation

Australia: how do we transform the organisation?

At the Australian Retail Council's 2026 events, AI was firmly on the agenda, covering everything from customer experience and merchandising to supply chain, workforce and operations.

But much of the leadership conversation came back to a familiar question: How do we introduce AI into the organisation we already have?

How do we get people to use it? Redesign roles? Govern it? Connect it to existing data? Move beyond pilots? And, ultimately, turn it into commercial value?

That makes sense because Australia has a large base of established businesses with existing systems, operating models, workforces and customer relationships.

So for many of them, AI is first and foremost an organisational transformation problem. The technology is moving incredibly quickly while organisations move much slower.

Singapore: what becomes possible if AI is part of the infrastructure?

This week, I've been at Tech Week Singapore, including Big Data & AI World Asia.

There is much more discussion around AI infrastructure, data, agents, automation, new architectures, AI-native operating models and what becomes possible when these capabilities are built into the organisation from the start.

So the question shifts to: What becomes possible if AI and data are fundamental to how the business operates?

Part of that is simply the nature of the events, geography matters as well. Singapore has deliberately positioned itself as a regional technology and AI hub, and many companies use it as a base for operating across Southeast Asia and the broader APAC region.

That creates a different commercial context. A company operating from Singapore isn't necessarily thinking about Singapore alone. It may be thinking about Indonesia, Thailand, Vietnam, Malaysia, the Philippines and beyond.

Which shifts the focus to: How do you build AI capabilities that work across fundamentally different markets?

Why won't AI make Asia one market?

One of the points that came through strongly in the Singapore sessions was that Asia isn't one market.

Different languages. Different consumer behaviours. Different platforms. Different data. Different regulations. Different levels of digital maturity.

AI doesn't remove those differences. If anything, it makes understanding them more important. An AI system that works in one market doesn't automatically become a universal operating model.

So I think the next phase of AI moves to: How does AI behave in this particular commercial environment?

And that brings me to what I think is the more interesting gap.

What is the missing layer between technology and commercial impact?

There are really three conversations happening.

Technology: What models, platforms, agents and data capabilities are available?

Transformation: How do we change our people, processes and operating models to use them?

Commercial: What actually changes about how customers discover, evaluate, choose and buy?

The first two conversations are well underway. The third is where I think things get much more interesting.

Because AI doesn't just change how companies work, but also how markets work.

What changes when AI moves from employee to customer?

Most enterprise AI today is still focused on helping employees do things better. A marketer analyses data with AI. A salesperson prepares for a meeting. A customer service agent gets assistance. A store manager receives a recommendation. The human is still making the decision.

But increasingly, AI is also entering the customer's decision process. What happens when someone doesn't search Google? They ask an AI agent to research the market. The agent compares products. It filters the options. It makes a recommendation, while the customer simply accepts it.

The growth question then changes to: How does an AI system decide whether our brand is relevant in the first place?

That's a very different problem.

Where does agentic commerce come in?

Search changed discovery. Social changed distribution. Marketplaces changed comparison. Recommendation engines changed personalisation. Now, AI agents have the potential to change selection itself.

The customer may no longer be the only entity evaluating the available choices. Their agent may do much of the work first.

That means brands need to be understood by machines as well as people. Their products. Pricing. Availability. Reputation. Differentiation. Policies. Proof points. Data. Relevance to a particular need. All of these can become inputs into machine-mediated selection.

Which raises a question I think every growth leader will eventually need to answer: If an AI agent were choosing between us and our competitors, what would make it select us?

What happens when capabilities move into the customer journey?

This is what the contrast between Australia and Singapore has made clearer for me.

The Australian conversation is largely about transforming the organisation: leadership, capability, operating models, adoption and governance. The Singapore conversation is much more focused on what AI makes possible: infrastructure, new capabilities, regional scale and new ways of operating.

Between those conversations sits another question: What happens when those capabilities move outside the organisation and into the customer journey?

That's where I think agentic commerce becomes important.

Why is "AI marketing" the wrong frame?

And it's also why I'm increasingly sceptical of the phrase "AI marketing." It makes a fundamental shift sound like another channel, another platform or another way to produce content.

But, I think the change is much bigger. AI is changing how companies operate, how decisions get made, how customers discover options, how products are evaluated and how choices are made.

This changes growth mechanics which then leads to the question: What happens to growth when intelligence becomes part of the infrastructure through which businesses and customers make decisions?

That's the question MARK measures — whether AI selects your brand and products, and what evidence appears to drive that selection.