AI is everywhere, but where's the return?

AI is attracting extraordinary levels of investment. But in conversations with customers across ANZ, the real question for most businesses is a simple one: when do we actually see a return on that investment?

For most organisations, the answer is not found in model benchmarks or funding announcements. It appears in the daily experience of their organisations. Is it removing friction from everyday work? Are teams getting through work faster? Are systems running more smoothly?

Moving beyond the excitement phase

The first wave of AI adoption was driven largely by possibility. What could these systems achieve? Pilot projects were launched across departments, often with genuine enthusiasm.

Now the environment is more demanding. Economic uncertainty persists, growth forecasts remain cautious, and operational costs are under closer scrutiny. In that context, experimentation alone is no longer enough. Businesses need AI to deliver measurable operational value.

That matters because many organisations are already operating under strain. Hybrid working has added layers of complexity, technology stacks have expanded over time, and skills gaps remain difficult to close.

The result is digital friction. This is everyday dysfunction that interrupts workflows and drains productivity.

Our latest research in Australia highlights how widespread this problem has become. 81 percent of employees say they lose time each month because of dysfunctional IT. On average they lose 1.3 workdays to digital friction, while more than half report delays to critical operations or projects as a result.

In this context, AI should not add complexity. It should remove it.

The organisations seeing the strongest ROI from AI are focusing on a clear objective. They want to reduce friction in how work actually gets done.

From reactive IT to proactive and autonomous operations

Many IT operations still follow a reactive model. A system fails, a ticket is opened, and teams respond. The process is familiar but inefficient. Teams spend large parts of their time responding to problems instead of preventing them.

AI enables organisations to break this cycle. In fact, 45% of Australian respondents believe AI can reduce digital friction, while 53% are open to it managing routine IT tasks such as troubleshooting and password resets.

AI enables organisations to move beyond this cycle.

By analysing operational signals across devices and systems, AI can identify patterns that signal emerging issues before they disrupt users. It can also generate and execute automations that resolve those issues earlier.

The next step is autonomous IT operations. AI systems can generate remediation steps, create automations, and resolve recurring issues automatically. Routine problems such as configuration errors, system slowdowns, or device inconsistencies can be detected and corrected without manual intervention.

It reduces downtime, lowers support workloads, and allows IT teams to focus on strategic work instead of repetitive troubleshooting.

The role of generative and agentic AI

Generative AI can automatically create documentation, troubleshooting guides, remediation plans, automation workflows, and even code.

Another development gaining attention is agentic AI.

AI agents can operate within defined environments. They can trigger automations, coordinate responses across systems, and resolve IT issues as they occur.

For organisations facing shortages of experienced technical professionals, this capability is particularly valuable.

Making work feel more manageable

These capabilities matter because modern organisations are deeply connected. Devices, systems, and people interact constantly. While this connectivity enables scale and flexibility, it also increases the likelihood of friction.

AI can reduce that strain in practical ways. In environments besides IT support, such as manufacturing plants, logistics hubs, or field service operations, workers frequently encounter technical issues that interrupt productivity.

In those moments, AI systems can surface relevant knowledge instantly, generate troubleshooting steps, or trigger automated remediation processes.

Each of these improvements helps employees resolve issues faster and with less disruption.

Work begins to feel more manageable.

A practical path forward

As AI becomes more embedded in daily operations, confidence will matter as much as capability. Organisations must be clear about how these systems are used, where human oversight remains essential, and how the technology is designed to support people in their work.

The next phase of AI adoption will be defined less by experimentation and more by integration. The organisations that gain the most will focus on outcomes. These include reducing friction, automating routine work, and helping employees operate more effectively.

When AI becomes a practical everyday helper across an organisation, adoption follows naturally. As the excitement around AI investment settles, usefulness will ultimately decide what stays. The technologies that endure will be those that quietly help organisations run better every day. That is where the ROI of AI will be found.