SYDNEY — OpenAI and Anthropic have urged Australia to reconsider a ban on allowing them to use the country's creative content to train their AI models, saying a limited exemption could protect artists' rights while supporting billions of dollars in prospective investment, according to Reuters.
The intervention puts two of the world's most valuable AI companies on a collision course with Australia's creative industries, and it comes as governments worldwide wrestle with the same question: whether AI developers should be able to train on copyrighted work without consent or payment. Readers tracking AI news will recognize the pattern — the same fight is playing out in courtrooms and legislatures across the US, UK and EU.
What the Companies Are Proposing
In submissions to a parliamentary inquiry examining Australia's AI laws, both companies acknowledged the existing ban but called for it to be relaxed. Anthropic, which owns the Claude chatbot, said in its submission that broad copyright exceptions have not received widespread support and have already been ruled out by the government.
As an alternative, Anthropic proposed that the government allow AI model training through a narrow form of conditional exemption. If Australia requires "investment or other conditions designed to support the future of Australian creators and cultural endeavours," the company said it is "open to considering those conditions."
OpenAI, for its part, called for "a balanced copyright framework that allows models to learn from publicly available information" while offering rightsholders opportunities to collaborate. Both companies framed their proposals as a middle path between the all-or-nothing positions that have defined the copyright debate elsewhere.
Australia Has Already Ruled Out a Blanket Exemption
The requests face a skeptical audience. Australia has said it will roll out a broad set of rules on AI and data centres from next year, but it has already ruled out an exemption to copyright laws that effectively stops the mostly US-domiciled AI companies from carrying out model training in the country.
The stakes for Australia are considerable. Economists cited by Reuters estimate that AI-related investment could be worth hundreds of billions of dollars by the end of the decade, and the government is simultaneously trying to position the country as a data centre hub while protecting one of its most culturally and economically significant export sectors: creative content.
The companies' submissions add to pressure on the government as it weighs the interests of AI developers against those of authors, publishers and other creators, who have resisted exemptions allowing their work to be used to train commercial models without consent or payment.
A Global Regulatory Race
The submissions land at a delicate moment in a worldwide contest over AI rules. The companies said countries around the world are still working out how to regulate artificial intelligence while trying to attract investment that economists say could be worth hundreds of billions of dollars by the end of the decade. Australia's decision will therefore be read less as a local copyright dispute and more as a signal of whether a mid-sized creative economy can hold the line on payment for content while still courting the AI industry's capital.
The Infrastructure Connection
Notably, both companies tied their copyright arguments to major infrastructure commitments in Australia. OpenAI has signed an offtake agreement with Australian data centre developer NextDC for a planned data centre in Sydney, while Anthropic was named just last week as a partner for a data centre in Queensland.
That linkage is the clearest evidence yet of the quid pro quo Anthropic's submission gestures toward: AI firms are prepared to spend heavily on Australian computing infrastructure, and they want copyright reform as part of the bargain. Critics will read the same dynamic differently — as wealthy foreign companies leveraging investment promises to extract policy concessions.
Australia is not an isolated case. The UK has spent the past two years fighting over nearly identical proposals, and the outcome of Australia's inquiry is being watched as a template for how mid-sized creative economies might respond to AI training demands. As we covered in recent weeks, Anthropic's expanding Australian footprint — from its Queensland data centre plans to its regional investment push — has made the country a testbed for how AI investment and AI policy develop together.
What Happens Next
Australia's Joint Select Committee on Artificial Intelligence is due to deliver its report in November. Until then, the government faces a well-funded lobbying push from two of the industry's biggest players, set against a creative sector that has shown no appetite for unpaid use of its work.
The companies' conditional-exemption proposal — trading copyright exceptions for creator-support investments — is the most concrete framework any major AI lab has put on the table in Australia. Whether Canberra treats it as a genuine compromise or a negotiating position will shape not only Australia's AI industry, but the playbook every other government reaches for next.
---
Stay Ahead of AIGet the latest AI news, analysis, and breakthroughs — all in one place.
Read more AI news →
