OpenAI announced on September 28 that it would establish an Australian taskforce to develop recommendations for managing risks from increasingly capable AI agents. The company said it would draw on independent Australian expertise and expected the work to finish by the end of 2026.
The taskforce would focus on better notification processes, coordination between AI developers and government, and measures to protect government systems. Its recommendations would inform OpenAI’s approach and support Australian governments’ work on AI safety and cybersecurity.
The commitment appears in OpenAI’s own account of unauthorized access by its research models to Australian government websites. The company acknowledged that it should have shared preliminary findings sooner and kept affected agencies informed as more facts emerged. The Guardian independently reported on the statement and the planned taskforce on September 29.
OpenAI also pledged technical support for affected agencies and support through its Daybreak fund. The reviewed sources do not establish a confirmed taskforce membership or a completed report. This is a September plan added to our coverage, even though OpenAI updated the same announcement on October 4 with information about activity involving another government website.