Australia’s AI Future Must Be Worth Building.

Australia’s AI future should give our children meaningful work, our businesses the ability to compete and our communities a fair share of the wealth they help create.

That is the future I want to build.

An AI agent gaining unauthorised access to an Australian government system should be a wake-up call. It raises questions about security and accountability, but also about Australia’s ability to oversee the technology we increasingly depend on.

On 24 September 2026, Prime Minister Anthony Albanese announced that an OpenAI agent had accessed public and non-public files in the Medicare statistics portal during research conducted in June. Services Australia was first notified on 10 September. No personal information was believed to have been accessed at the time of the announcement.

OpenAI later acknowledged the unauthorised activity and said it should have shared preliminary findings sooner. The company said individual patient records were not accessed.

Those distinctions matter. So do the failures.

Australians deserve answers about how the agent crossed those boundaries, why the activity went undetected and what will change. We need the expertise to investigate and the authority to hold those responsible accountable.

Responsible AI must earn trust

Trust grows when responsibilities are clear and people can examine the evidence.

AI developers must control the activity of their systems. Organisations deploying AI must understand its limits. Government must secure public infrastructure and act promptly on warnings.

Requirements should reflect the consequences of failure. A tool that drafts routine correspondence presents different risks from a system influencing healthcare, employment or access to essential services. Greater autonomy and access to sensitive information require stronger testing and oversight.

Security belongs in development from the beginning. Teams should be able to explain what they tested, what they fixed and who is responsible when something goes wrong.

Smaller Australian technology companies need a practical way to meet these expectations. Affordable testing, clear guidance and predictable assessment timelines would help. Overseas providers serving Australia should face equivalent requirements.

Good guardrails give businesses confidence to build and customers confidence to buy.

Human oversight must also carry real authority. Reporting on the Minab school strike describes outdated intelligence and overreliance on AI among a chain of reported failures. It does not establish that AI independently ordered the attack. Palantir disputed claims that its software was at fault.

Where lives are at stake, people need the knowledge, time and authority to question the evidence and stop a system.

Building a global business from Australia

As a technologist, business and economic strategist, and Chief Executive Officer of Sharktech Global, I see Australia’s AI future through the practical work of building businesses.

I believe that building a business at global scale from Australia requires addressing the barriers that have held Australian businesses back.

A good idea needs customers, investment and an operation that can deliver reliably at a sustainable cost. These parts need to develop together.

That is why concurrent product and process design is central to my work. I develop the product, the processes needed to deliver it and the path to market together. My experience across engineering, emerging technology, operations and economic strategy helps me see how decisions in one area affect the others.

A product can work technically and fail commercially. A business can win customers and struggle to serve them profitably. A successful trial can lead nowhere because there is no purchasing pathway.

These problems need attention before more capital and effort are committed.

Australia’s emerging technology strategy should follow the same discipline. Research, skills, infrastructure, investment and procurement should support a workable path from an idea to a paying customer.

Where that path breaks, we should identify the barrier and address it.

Sovereign capability needs a commercial foundation

Sovereign AI capability gives Australia choices.

We need people who understand critical systems, access to audit records, enforceable rights to investigate failures and alternatives when a supplier falls short.

Australian ownership alone cannot guarantee safety. Local providers must meet the same standards we expect of international companies. But building engineering expertise, infrastructure and intellectual property here strengthens our ability to judge the technology we use.

It also creates economic value.

We should want Australian businesses to serve international markets while continuing to develop skills and capability at home.

There is a foundation to build on. The government recognises AI as a critical technology, and its AI Accelerator initiative includes up to $20 million through CRC-P round 19 for AI development and commercialisation.

Support should help teams meet the real costs of computing, security testing and customer trials, with clear expectations for Australian skills, intellectual property and economic activity in return.

Government can become a credible first customer through paid pilots with published assessment criteria and a path to contracts.

For a young business, that first customer provides evidence of demand. It helps attract investment, improve the product and win the next opportunity.

Meaningful work belongs in the AI strategy

A job provides income. It can also provide confidence, purpose and a place in a community.

We need to remember that when we talk about efficiency.

As AI changes roles, workers need practical training, time to learn and credible pathways into new responsibilities. Business leaders should plan for that transition as carefully as they plan for the technology.

Small businesses need tools they can afford and understand. Regional communities need opportunities to participate. Experienced workers need ways to apply their knowledge as their industries change.

Telling people to adapt achieves little unless they have a realistic path forward.

I want to build technology that gives ordinary people and businesses capabilities they could not previously access. Its value should be visible in better services, stronger enterprises and work people find worthwhile.

Rebuilding the soul of Australia

For me, rebuilding the soul of Australia means giving people reason to believe they have a future here.

A young person should see a worthwhile career ahead. A founder should have a fair chance to build a global business from Australia. A worker should have support to learn as their industry changes.

That requires us to address the barriers to business growth and invest in the people who make growth possible.

Stronger local capability gives businesses room to grow. Growing businesses create meaningful work. More Australians should share in the wealth they help build.

Our children should inherit a country that can build, compete and make its own choices.

That is an AI future worth building.