Australia’s role in the frontier world of artificial intelligence was the focus in Canberra last month, with two events bringing together hundreds of defense, diplomacy, development, and industry experts. The Australian Strategic Policy Institute (ASPI) gathered the national security and defense community for an AI Masterclass on September 14. The next day, the Development Intelligence Lab’s Situation 2026 explored how AI can solve Indo-Pacific development challenges.
The events revealed a country grappling with what it means to be a middle power in the age of AI. As ClearAI founder Jisoo Kim put it, at a time when AI is developing at breakneck speed and reordering economies and societies, “Where in this ecosystem, where in this global economy is Australia going to sit?”
Judging from opinion polling, many Australians would probably prefer that AI had never been invented. But now that the cat is out of the bag, Australia needs to think about what it means for its security and prosperity, and the risks and opportunities it brings.
As a middle power, with a range of assets and resources, Australia is not powerless in the face of these changes. It has some capacity to influence the new world that AI is making.
Assistant Minister for the Digital Economy Andrew Charlton addressed this openly. Asking how middle powers manage AI, he identified four plays: alignment with a superpower, leverage in the value chain, competition through different models and capabilities, and sovereign capability. Charlton stated that Australia is pursuing all four simultaneously.
Shadow Minister for Industry and Sovereign Capability Andrew Hastie broadly agreed, particularly emphasizing the U.S. relationship as Australia’s “line to sovereignty.” Hastie suggested using Australia’s relationships in Washington to encourage a hyperscaler to invest in Australia.
Without data centers, frontier labs, and talent, Australia remains dependent on what comes through submarine cables, and this may make Australia a “supplicant state” – with Hastie using the analogy of the nuclear race, where once the U.S. convinced Australia to withdraw, it has been reliant on Washington ever since. He gave a sense of the stakes for Australia: countries with AI will pull ahead and Australia will fall further behind.
According to Minister for Industry and Innovation Tim Ayres, the government’s “aim is not to do everything, but to make ourselves the indispensable country.” That is, to find what Australia can excel at in this new ecosystem and make it the country’s comparative advantage.
Australia brings assets in policy expertise, talent, and trust. Ayres also highlighted capabilities in chip packaging, quantum computing, and specialized datasets, alongside institutional advantages such as firms and unions working together in the national interest.
Throughout the discussions, there were multiple calls for Australia to work with other countries including the United Kingdom, Canada, Japan, South Korea, and India.
As the American Enterprise Institute’s Ryan Fedasiuk put it, middle powers have a path forward if they “conceive of AI as a global megaproject which everyone can contribute to.” Importantly, “Sovereignty doesn’t mean autarky. Sovereignty means being able to choose your own dependencies rather than having them foisted on you.”
Aditya Maheshwari of the Indian Institute of Management Indore outlined how middle powers could energize their own ecosystems through talent, research, joint projects, and shared computing facilities, rather than competing with the United States and China on frontier models.
Australia also has considerable opportunities to shape international standards. Australian Signals Directorate Director-General Abigail Bradshaw argued that having a place in the AI industry strengthens Australia’s sovereignty and gives it a voice in global standards.
Australian AI Safety Institute head Kate Conroy highlighted Australia’s ability to punch above its weight internationally, drawing on its experience in developing standards adopted globally.
Australia’s Ambassador for Cyber Affairs and Critical Technology, Jessica Hunter, argued that Australia has the assets, relationships, and agency to be a shaper of the AI world rather than merely a consumer.
It’s interesting that this proactivity is being actively encouraged by frontier labs, including OpenAI Vice President Ann O’Leary and Anthropic Special Envoy and former U.S Ambassador to Australia Jeff Bleich.
An important element of being an AI middle power is to think about how it contributes to development outcomes. As Minister for International Development Anne Aly put it, “The value of technologies lies in the way that people’s lives are improved.” The challenge is “to ensure the benefits of emerging technologies are shared widely rather than used to widen inequalities.”
Through this lens, Australia’s Department of Foreign Affairs and Trade sponsored a competition on what Indo-Pacific development challenge AI could solve in the next five years, gathering more than 100 proposals. From disaster warning systems to healthcare and protecting Pacific maritime zones, there are countless ways Australia can use AI to contribute as a development partner.
Times of change are generally good for middle powers, enabling them to play an entrepreneurial role. These discussions demonstrated the benefits of Australia inserting itself into the AI economy, shaping global safety standards and enabling positive diffusion of technology across the region.