A bipartisan parliamentary committee has recommended placing independent experts inside leading AI companies, giving them access to the companies’ systems and a duty to report safety findings and incidents to government.
In early advice delivered to the Prime Minister on 12 October, the committee also proposed giving the Australian Signals Directorate early access to new frontier models as a condition of training them in Australia. Developers would face clear obligations to monitor risks and promptly report incidents. Other recommendations include reserving domestic computing capacity for national-security and research purposes, reviewing government data protections and reconsidering copyright settings.
These proposals are not law. The Joint Select Committee on Artificial Intelligence is due to deliver its final report on 30 November, while the government has indicated that legislation establishing national standards could be introduced before year’s end.
The significant shift is from disclosure after something goes wrong to continuous, independent access to the systems creating the risk. Embedded monitors could test safety claims, examine unexpected agent behaviour and report directly to government rather than relying entirely on providers’ internal judgments.
For lawyers advising AI developers and users, that model would make audit access, regulator-facing records, incident thresholds and reporting responsibility central contractual issues—not merely voluntary governance features.
Sources: ABC News; parliamentary inquiry
Leave a Reply