Artificial Intelligence (AI) is coming up in many community conversations I’m having at the moment – how kids are using it, how it’s changing people’s jobs, dystopian fears about the future, or why our future is being shaped by a handful of foreign billionaires.

The productivity benefits of AI are immense, but so are the risks.

When other potentially harmful technologies have developed (like cars and the internet), it has taken a while for safety regulations to catch up – law-making moves slowly.

But if we don’t make decisions about how we want AI to fit into our future, those decisions will be made for us on the other side of the world.

My Curtin AI survey showed that people are four times more likely to be concerned about AI than optimistic about it. Eight in ten don't think the government is doing enough. Eight in ten don't believe current policies will ensure AI benefits Australians rather than overseas tech companies.

So, how should our government respond in this context?

After consulting with my community and 30 Australian and international experts from civil society, industry and academia, I have released an AI Discussion Paper: Shaping the future of AI for Australia. It proposes 18 practical policies across five goals.

We need to set up the structures — including properly funding the Australian AI Safety Institute, which currently receives less than $8 million a year while its UK equivalent receives fifteen times that amount.

We need to harness opportunities through a National AI Missions program that backs Australia's genuine strengths: natural disaster prediction, agricultural resilience, health research using our Medicare data, and the clean energy transition.

We need to address current harms — establishing a digital duty of care, extending the under-16 social media ban to AI companion apps that simulate emotional intimacy with children and confronting the tsunami of AI-generated deepfakes and disinformation.

We need to prepare for emerging risks by stress-testing our laws and requiring frontier AI developers to be transparent about their systems.

And we need to share the benefits — converting the government's voluntary data centre expectations into binding obligations, and taxing international tech companies so the extraordinary economic value they generate in Australia flows back to Australians.

In a rapidly changing world, we can’t afford to wait and see. I will keep pushing the Government to set guardrails that are consistent with our values so we can take up AI with optimism, and everyone can share in the benefits.

Click here to read my AI Discussion Paper: Shaping the future of AI for Australia.