Anthropic moves on Australia as AI race with OpenAI heats up

Anthropic is stepping up its push into Australia, signing a memorandum of understanding with the federal government as competition with rivals such as OpenAI intensifies and billions flow into local AI infrastructure.

The agreement will see the AI company collaborate with Australia’s AI Safety Institute, share research on emerging model capabilities and risks, and track how AI is being adopted across key industries.

It marks one of the first major initiatives under the government’s National AI Plan and signals a deeper alignment between policymakers and frontier AI developers.

Under the deal, Anthropic will work with Australian institutions on safety research and provide data from its Economic Index to help monitor AI’s impact across sectors including natural resources, agriculture, healthcare and financial services.

The company is also exploring further investment in Australia, including data centre infrastructure, as it looks to build out local capacity and meet growing demand from enterprise and government customers.

The move comes as Anthropic prepares to open a Sydney office, its fourth in Asia-Pacific, following Tokyo, Bengaluru and Seoul.

At the same time, the company is backing Australia’s research and startup ecosystem with targeted funding.

Anthropic has committed $3 million in Claude API credits to four institutions — the Australian National University, Garvan Institute of Medical Research, Murdoch Children’s Research Institute and Curtin University — supporting projects in areas including genomics, precision medicine and computing education.

It has also launched a new program offering up to US$50,000 in API credits to deep tech startups working in areas such as drug discovery, climate modelling and medical diagnostics.

The expansion reflects growing demand for AI across Australia and New Zealand, with both markets ranking among the highest globally for usage of Anthropic’s Claude platform relative to population.

For the federal government, the partnership is part of a broader push to position Australia as a trusted hub for AI development, with a focus on safety, infrastructure and economic impact.

For Anthropic, it is a strategic move to deepen its presence in a market where enterprise adoption is accelerating and policy frameworks are still being defined.

The timing is notable. As OpenAI continues to outline its own ambitions in Australia, Anthropic’s latest moves signal a more direct play to embed itself within the country’s research, policy and infrastructure ecosystem.

The infrastructure push is also being matched by private capital.

AI infrastructure startup Firmus is preparing for a potential IPO at a valuation nearing $7 billion, with Nvidia reportedly doubling down on its investment as part of a $1 billion-plus pre-listing raise.

The company is building out what it describes as “AI factories” in Australia, including a major deployment in Melbourne backed by tens of thousands of Nvidia GPUs and long-term contracts with global technology partners.