Anthropic Proposes Opt-Out Model for AI Training on Australian Copyrighted Content

Anthropic's Proposal for Content Access

AI company Anthropic has urged the Australian government to consider conditional approval for AI models to train on Australian copyrighted works under an opt-out system. This proposal comes after the government ruled out a blanket text and data mining exemption for AI companies. Anthropic acknowledges the dual nature of AI development, citing potential economic transformation and productivity gains alongside national security and critical infrastructure risks.

Public Broadcasters' Opposition

Australia's public broadcasters, ABC and SBS, have strongly criticized AI firms' current practices. They advocate for strict new rules, demanding that AI companies adhere to the same copyright, defamation, and privacy regulations as traditional media outlets. The ABC warned that unregulated AI could lead to the 'cannibalisation' of mainstream news, while SBS highlighted challenges to the sustainability of the Australian news industry and the erosion of information integrity.

Call for Inclusion in News Bargaining

Both ABC and SBS have suggested that AI companies be included in Australia's news bargaining incentive. This would require AI firms to enter commercial agreements with large media organizations to support journalism, similar to existing arrangements for other digital platforms. They argue that the current unregulated status of AI-based services provides them with more favorable treatment than established Australian media, exacerbating an unequal playing field.

Upcoming Parliamentary Inquiry

A joint select committee on AI is scheduled to hold a series of hearings in Sydney. Executives from Anthropic and OpenAI are expected to attend. This inquiry will address various aspects of AI development and regulation, including access to copyrighted materials for model training and the broader implications for Australian industries and society.