Anthropic pushes for opt-out model for Australian content as ABC warns of ‘cannibalisation’ of news

Anthropic’s chief executive, Dario Amodei. The AI company has been pushing for ‘conditional approval’ for big tech to train its models on Australian copyrighted works under an opt-out model. Photograph: Andrew Thomas/NurPhoto/Shutterstock View image in fullscreen Anthropic’s chief executive, Dario Amodei. The AI company has been pushing for ‘conditional approval’ for big tech to train its models on Australian copyrighted works under an opt-out model. Photograph: Andrew Thomas/NurPhoto/Shutterstock Anthropic pushes for opt-out model for Australian content as ABC warns of ‘cannibalisation’ of news Maker of Claude claims AI could transform economy but ABC and SBS say the technology should be subject to media regulations Get our new political email, free app or daily news podcast AI giant Anthropic has urged the Albanese government to consider giving “conditional approval” for big tech to train its models on Australian copyrighted works under an opt-out model, after conceding it won’t secure a blanket Copyright exemption. But Australia’s public broadcasters, the ABC and SBS, have strongly criticised AI firms, calling on the government to enact strict new rules to compensate media organisations and protect public interest journalism. In submissions to a federal parliamentary inquiry, the ABC demanded AI companies be subject to the same Copyright, defamation and privacy rules as media outlets, warning of a “cannibalisation” of mainstream news. The ABC and SBS also suggested the government include AI companies in the News Bargaining Incentive, which would require them to make commercial deals with large media outlets to support journalism. “As publishers, AI-based services are currently unregulated and receive more favourable treatment than established Australian media. This unequal treatment will accelerate the cannibalisation of the Australian news industry,” the ABC said in its submission. SBS said AI-powered large language models – such as ChatGPT or Claude – were “challenges to the sustainability of the Australian news industry and the erosion of information integrity and public trust through their use of news content without permission, attribution or compensation”. The joint select committee on AI will hold a series of hearings in Sydney next week, with executives from Anthropic to attend along with those from OpenAI, after it admitted one of its AI agents had accessed Services Australia data in June. In a newly published submission, Anthropic said there was “light and shade” to be considered around the development of AI. It claimed the technology could “transform the Australian economy, drive a new wave of productivity” and “solve some of the nation’s biggest challenges”, but also warned of “serious implications for Australia’s national security, the resilience of its critical infrastructure and the safety of its people”. The company talked up its plans for investing billions in Australia, including large new datacentres, and urged the government to support “frontier model training”. But one sticking point raised by AI companies has been access to Australian copyrighted works – such as news, books, music and other written material – to help train AI models. The Labor government has ruled out a so-called text and data mining exemption, with the prime minister, Anthony Albanese, promising “the strongest possible protection” for Australian creatives. But questions have lingered about other possible Copyright changes and the suggestion of an opt-out measure, allowing Copyright holders to choose whether their content can be accessed by AI companies. Anthropic’s submission called on Australia to “provide legal certainty under Copyright that brings AI developers to the table while advancing its national cultural interests”. “Anthropic accepts that broad exceptions from Copyright, such as a text and data mining exception, have not received widespread support and have been ruled out by the Government. As an alternative, we believe that the Government could permit AI model training through a narrow form of conditional approval,” the company submitted. It suggested the implementation of “do not crawl” signals through robots.txt files – which tell bots what they may and may not access – and said the company was “open” to supporting Australian creators. However SBS said that such an opt-out measure may not work. “Robots.txt mechanisms are routinely ignored or bypassed, and the blocking of ever-evolving training bots increases the burden on rightsholders to monitor and update protections, when the onus should be on AI companies to avoid wrongdoing,” the broadcaster said in its submission. The ABC complained that AI companies did not abide by the same rules as media outlets, calling the current system “anti-competitive”. Both broadcasters called for AI to be included in the News Bargaining Incentive – with which big tech platforms such as Facebook, Instagram, Google, TikTok and Snapchat have been forced to comply. Explore more on these topics AI (Artificial Intelligence) Anthropic OpenAI Australian politics Anthony Albanese news Share Reuse this content