AI in motion… moving between maturity modes

In assessing AI appetite across Australia, three modes of business maturity have emerged and will define progress in 2026.

Firstly, some organisations are feeling the pinch to run at 100 miles an hour at risk of feeling left out or being passed by their competitors.

But they don’t truly understand any key risks or obligations against legislation, nor have awareness of the need for – and the importance of – quality trusted data to yield the required results.

At the other end of the spectrum, some organisations are taking a full aversion to risk to having AI being used in the business from restricting full use of any GPTs or empowering teams to explore in a ‘lab’ environment the opportunities that it could present to them.

They are aware of the potential pitfalls but are missing out on efficiency gains, service improvement and innovation.

Then we have the middle section of that maturity curve – some organisations are dipping their toe in the water, cautiously experimenting to find efficiency gains without the awareness of what to do next or how to improve compliance or improve the quality of data. This is the ‘curious and open to explore’ group but they remain unclear on what comes next.

There is no single ‘right’ approach as each organisation must assess its own risk appetite, the perceived benefits as well as its data maturity and compliance and regulatory obligations.

According to Moxie Research – AI Outlook: Australia 2026 – enthusiasm for AI initiatives remains strong at the executive leadership team level but that doesn’t always translate into execution.

In other words, sentiment is strong but support is weak – 93% of leadership teams in Australia are in favour of AI yet only 62% are deliberate in driving organisational change from the top.

Within this context – and building on the point of self-evaluation – two major priorities will emerge over the next 12-24 months in Australia.

1. Building AI literacy, governance, awareness and safe usage first.

AI maturity is not about the technology but the people who are using it and underpinning safe exploration and adoption, which will lead to better decision making.

I’ve seen organisations invest in building education and awareness first so staff can use AI tools, including GPTs safely and confidently. That means understanding there are benefits but also acknowledging the moral, legal and ethical obligations.

According to Moxie Research – and in response to increased demand for education and governance – Australian businesses in 2026 will:

• Invest in foundational AI education and training: 72%
• Strengthen governance and ethics framework: 69%
• Focus on change management: 59%

Organisations must ensure their workforce recognises that publicly available GPTs still have a high error rate, however.

Using a personal example from last year, I was researching different motorcycles to buy.

I was new to this market. I’m not overly tall and I wanted something that wasn’t too big, too ‘hooliganish’ and no tiptoeing when stopping. It was alarming how many of the recommendations simply didn’t exist – motorcycle models were fused together from different brands or entirely fabricated.

If that can happen with something as straightforward as motorcycle shopping, imagine when AI is used inside a business without the right understanding or guardrails?

2. Getting organisational data to ‘good’.

This second priority and differentiator is the quality of data being trusted, reliable and secure, which I believe will be the real competitive advantage with AI.

With strong data quality, governance and security, businesses can pivot faster, scale use cases and extract the value from an AI system.

The alternative is that organisations end up ‘fixing on the fly’, creating technical debt or introducing new risks. We all know the saying, ‘garbage in, garbage out’ which applies even more so with AI.

Organisations that invest in data governance, data quality and securing their data will end up the ones who accelerate faster.

A combination of forces will be driving these priorities, notably:

• fear of being left behind plus the pressure to innovate safely.
• teams exploring efficiency gains to spend more time on value-add tasks and improving productivity through AI.
• aligning to regulatory and compliance obligations as AI safety, privacy and ethical use is now evident in a number of legislations and Acts.

As a result – and based on Moxie Research – Australian businesses in 2026 will prioritise:

• Selecting the right tech stack / infrastructure: 76%
• Evaluating data readiness to maximise AI: 72%

While the promise of AI is undeniable, the challenges are becoming clearer – and more complex – the closer businesses get to real-world deployment.

From a roadblocks standpoint, most organisations struggle with three ongoing roadblocks:

1. Data isn’t ready for AI. If the data isn’t trusted, reliable and secure the use case fails. AI is not magically going to correct poor data management and business processes, that is still on us.
2. A workforce that isn’t equipped with the literacy to safely use and understand its limitations can put teams into ‘spinning wheel mode’ unaware of what to do next.
3. Failure to engage trusted third-parties and understand what AI functionality is being developed in their roadmap. Plus, organisations not talking to each other or exposing themselves to the right events to build up their awareness.

Moving beyond that, there are a number of obligations that organisations have under The Privacy Act, The Spam Act, Do Not Call Register (DNCR) or information handling rules that remain unknown to organisations.

The main concern is that organisations aren’t investing time or seeking guidance in understanding their obligations. Information and data handling obligations existed way before AI came into the picture so freezing is not an option.

These are all risks that need to be mitigated.

But the problem with AI isn’t the technology, it’s the legal and ethical obligations that organisations have, from privacy and data leakage, accuracy and reliability, bias, fairness and impersonation and ownership.

Evaluating AI use cases, building business value

At this stage of the maturity curve, most organisations have stood up select generative AI (GenAI) use cases but have yet to tackle more complex work such as data foundations, operating models and governance frameworks.

In 2026 – and as outlined via Moxie Research – 71% of Australian organisations will identify ‘high-impact’ AI use cases, leading to 68% deploying small pilot AI projects.

While use cases exist for building in-house GPTs and LLMs, AI isn’t new as businesses have being using this technology for some time.

Remember Clippy, the animated, paperclip-shaped assistant introduced by Microsoft in Office 97 to provide user help? That was a form of AI attempting to predict what a user was going to do but it didn’t learn anything new and existed to only help improve productivity.

Productivity support is probably the most widely used use case today in the form of summarising documents, notably legislation which is a key task at RSPCA Queensland.

Yes, there is a little legal nerd inside me that enjoys reading this type of document but AI helps summarise areas where I don’t have the legal jargon or large amounts of time to sift through multiple legislations and Acts.

Based on Moxie Research, Australian businesses are enhancing employee experience through AI in the areas of improved productivity (52%) and workforce optimisation (49%).

Also, chatbots and virtual assistants play a very big part in most common use cases, although I don’t think enough of this is currently used in organisations. Contact centre and IT service desks should already be exploring or embedding this functionality to provide quicker support to customers – plus, analysing the information afterwards is an even better use case.

According to Moxie Research, organisations are leveraging AI to improve customer experience through:

• Chatbots: 58%
• Voice Assistance: 50%
• AI Personalisation: 50%

The use cases highlighted stand tall as the top AI solution priorities for Australian organisations during the next 6-12 months, based on my own experience and opinion.

In saying that, every enterprise application, frontline application, operational technology or reporting application has already embedded AI functionality into their platforms or have very near roadmap initiatives to embed AI.

Think SAP, Microsoft and ServiceNow as some of the major players that have already embedded co-pilots and predictive intelligence into their core application, alongside integrating AI into their platforms to improve triaging and response generations.

What is driving these priorities are technology vendors that are moving faster than organisations can receive and safely implement, though as we receive software updates from suppliers those updates are including AI functionality already.

At some point, there must be a leap of faith to start using some of that functionality.

It’s not our partners responsibility to wait for organisations to be AI safe and ready, they have their own competing priorities and release cycles to get out into the market with new AI bells and whistles.

Partners need to be proactive to engage their customers and provide as much education on what is embedded and how an organisation can gain value, along with any guidance and support.

It would be an amazing add-on service for a lot of our partners to take the lead on helping organisations build those foundations on education, innovation and governance.

Andy Anastasi is CIO at RSPCA Queensland. As part of Moxie Top Minds, Andy contributed to AI Outlook: Australia 2026 by Moxie Insights. Download the report here.