Anthropic and OpenAI ask Australian lawmakers to relax restrictions that prevent AI models from being trained on Australia’s creative or other local material. The companies make the request in submissions to a parliamentary inquiry into proposed AI laws, according to Reuters reporting.
The issue centers on a legislative approach that, as described by outlets, does not create a specific “text and data mining” exception for training. One outlet says the companies’ arguments also reference that, in October 2025, the Australian government indicated it would not add such an exception. Across coverage, the companies argue that allowing training on local content would improve AI systems’ development and capabilities. Other perspectives, reflected in the broader policy debate around copyright and content rights, relate to concerns about permission, licensing, and potential impacts on creators—though details of those concerns are not the focus of the provided excerpts.
