What Australia’s Anthropic deal reveals about tracking AI at work

• Anthropic expands in Australia through a data and AI partnership.
• Supports universities and startups under national AI plans.

Efforts are increasing in Australia to track how artificial intelligence is affecting jobs and industries, with the government working with AI firms to gather data on its use in the economy.

One such effort is a new agreement between Anthropic and the Australian government covering economic data sharing, AI testing, and research.

At the core of the agreement is access to Anthropic’s Economic Index data, which tracks AI use and its impact on jobs and productivity. The government plans to use it to understand where AI is being adopted and how it may affect workers.

The analysis will first focus on sectors that play a large role in Australia’s economy. This includes natural resources, agriculture and financial services.

Anthropic said its data shows Australians are using Claude in a range of tasks. Compared with other English-speaking countries, use in Australia appears more varied. People are applying AI in management, sales, business operations, and life sciences. Many also use detailed prompts for more complex work.

Safety work and national AI plans

Alongside data sharing, the agreement includes joint work on AI safety. Anthropic will share research on how its models behave, including their limits and risks. It will also take part in joint testing with Australia’s AI Safety Institute and work with universities on related studies. Similar arrangements are already in place with government bodies in the US, the United Kingdom, and Japan.

“Australia’s investment in AI safety makes it a natural partner for responsible AI development,” said Dario Amodei during a visit to Canberra, where he met with Anthony Albanese. “This memorandum of understanding gives our collaboration a formal foundation.”

The agreement also ties into Australia’s broader plan to expand AI use in the economy. Under its National AI Plan, the government wants to attract investment in data centres and improve access to computing power. It also plans to build skills for AI-related work. Rather than introduce new laws, it will rely on existing rules and voluntary guidelines on safety and privacy.

The company said it is exploring investment in data centre infrastructure and energy projects in Australia, in line with government expectations. The projects are expected to support the growing demand for computing resources needed to run AI systems.

Research funding and industry support

The partnership includes direct funding for research. Anthropic has set aside AUD$3 million in API credits for four Australian institutions: the Australian National University, the Murdoch Children’s Research Institute, the Garvan Institute of Medical Research, and Curtin University. These organisations will use AI tools to support work in areas like medical research and computer science education.

Some of the projects focus on healthcare. Researchers at the Australian National University are using AI to analyse genetic data linked to rare diseases. At the Garvan Institute, teams are working on ways to connect genetic variation with how diseases behave in specific cell types, which could help identify new treatment options. Work at the Murdoch Children’s Research Institute includes studies on childhood conditions.

There is also a focus on skills and training. The agreement includes plans to support AI education, with universities using tools like Claude in teaching and research.

Beyond academic research, Anthropic is looking to support startups in Australia. The company plans to offer API credits of up to US$50,000 to selected firms working in areas like drug discovery, climate modelling, and medical diagnostics, a programme aimed at helping smaller companies build products using AI tools.

The steps come as Anthropic begins to expand its presence in the Asia-Pacific region. The company is preparing to open an office in Sydney and build a local team.

While Australia does not yet have AI-specific laws, the government’s approach points to gradual adoption with oversight. By working with AI companies, it wants to gather data and test risks. It also wants a clearer view of how the technology is being used in different sectors.

The agreement with Anthropic adds another layer to this approach, linking policy goals with practical use cases in research and education.

Want to experience the full spectrum of enterprise technology innovation? Join TechEx in Amsterdam, California, and London. Covering AI, Big Data, Cyber Security, IoT, Digital Transformation, Intelligent Automation, Edge Computing, and Data Centres, TechEx brings together global leaders to share real-world use cases and in-depth insights. Click here for more information.