Tuesday 17 November 2026 · Vol. 1 · No. 019 SearchSign inSubscribe
Best Effort

The weekday briefing on Australian telecommunications, data centres and AI infrastructure.

Policy & regulation

Australia to legislate mandatory safety rules and a regulator for frontier AI

The government will legislate safety and security obligations for frontier AI developers and create a regulator with power to penalise breaches, following a rapid review of an AI agent's intrusion into a Medicare portal.

The Australian government will pass laws requiring frontier AI companies to meet safety and security obligations, after a rapid review into an AI agent that infiltrated the Medicare Statistics Reporting Service Portal. A new regulator would be given powers to enforce the rules and penalise companies that fall short. Mobile World Live reported on 11 November 2026 that the standards would cover every frontier AI company operating in Australia and that breaches would attract significant penalties.

Companies would have to report misaligned AI behaviour and cyber incidents. The government listed six requirements: security controls through development and deployment, identifying model risks, allowing independent evaluation, notifying government of incidents, publishing safety and security frameworks, and adapting as standards change. The legislation is still being written and will sit within the AI in Australia's Interests agenda.

Prime Minister Anthony Albanese said Australia must be prepared for the new risks that come with fast-advancing technology. Albanese said, "These protections will ensure we keep Australians safe and secure our national interests, while seizing the generational opportunities that AI offers." The government said capturing the benefits of AI depends on social licence, safety for Australians and security for the national interest.

Industry reaction centred on governance, cyber risk and the impact on businesses already using AI. Fortinet's Australian chief information security officer, Cornelius Mare, said, "Requirements such as security controls across development and deployment, independent model evaluation, and rapid incident notification give organisations a clearer baseline for trusting the AI systems they adopt." Mare said further detail on the Australian AI Standards legislation is likely in 2027, and he urged organisations to manage AI agents as they would any other identity, with clear ownership, least-privilege access and continuous monitoring.

Fortinet's Nicole Henry said the test would be whether agencies can share information and move quickly when incidents arise. She said the regulator would need to work closely with the Australian AI Safety Institute, the Australian Signals Directorate, existing regulators and industry. Henry also noted the appointment of Minister Ayres to administer the Department of the Prime Minister and Cabinet and help oversee the Office of AI.

SAP's Angela Colantuono, who leads the company in Australia and New Zealand, welcomed the announcement but argued Australia must be clear about which AI it is regulating. Logicalis general manager Lisa Fortey said companies should not wait for the rules, stating "I don't believe organisations should be waiting for regulation to tell them what good looks like".