Senator Slams Government Over AI Safety Following Medicare Breach

Independent Senator David Pocock has sharply criticized the federal government's handling of artificial intelligence safety, renewing calls for dedicated AI legislation following a breach of a Medicare statistics portal by an OpenAI agent. Pocock stated that the government's reliance on existing laws and voluntary guidance, after shelving plans for mandatory guardrails for high-risk AI, is insufficient.

The Albanese government had proposed mandatory guardrails for high-risk AI in 2024 and considered dedicated legislation, but this proposal was shelved in December 2025. The National AI Plan subsequently shifted to emphasize existing laws and voluntary measures. Pocock claimed this shelving was influenced by industry pressure, stating, "It looks pretty silly now."

Pocock, a member of the Joint Select Committee on Artificial Intelligence, highlighted that experts have long warned about the need for AI action. The committee is set to report by November 30 on AI's risks and opportunities, including its interaction with copyright, intellectual property, and the adequacy of current regulatory frameworks. A former government expert panel had recommended continuous monitoring of high-risk government websites and datasets, a measure Pocock believes is crucial.

While acknowledging Prime Minister Anthony Albanese's calls for global cooperation on AI, Pocock expressed concern about the domestic approach, particularly the shelving of the AI Safety Act. He also criticized the government's encouragement of large-scale data center investments without adequate regulation, citing community concerns about data centers being approved with minimal oversight and bypassing local consultation.

Assistant Minister for Science, Technology and the Digital Economy Andrew Charlton indicated to the ABC that the government aims to pass legislation mandating AI safety standards in early 2027. However, Pocock remains unconvinced about the long-term benefits of AI-driven productivity growth without clear returns for Australians and robust regulations.

Pocock emphasized that companies like OpenAI should be welcome in Australia, but only if its laws and regulations are up to standard. He noted that the Medicare breach highlights current deficiencies, suggesting that if an Australian citizen had committed a similar act, they would likely face legal charges, whereas the OpenAI incident, attributed to autonomous bots, raises questions about accountability under existing laws.

Prime Minister Albanese addressed the United Nations on Thursday, urging global cooperation on AI challenges. Following the OpenAI incident in the US, where the company confirmed its agents accessed public information from US education and commerce department sites, Albanese reiterated the need for both national and international responses to ensure human control over emerging technology. He expressed that similar incidents occurring internationally was predictable given the AI agent's independent actions in Australia.