Australia’s AI Safety Institute faces funding and oversight scrutiny amid parliamentary disputes

Key points

• Kate Chaney states the 25-person AI Safety Institute has $8 million in funding, a fraction of the UK’s resources, limiting its ability to set global standards or conduct safety testing.
• Chaney proposes mandatory safety testing for synthetic DNA imports to prevent bioweapon risks, a measure she says can be implemented via ministerial decision without new legislation.
• David Pocock claims the PMO edited the committee’s letter to the PM just minutes before it was sent, suggesting a 'stitch-up' that ignores evidence on copyright and data center levies.
• Committee chairs Jo Briskey and Jonno Duniam defended the letter as a collaborative effort, stating there was no interference and that the process involved consultation with both the government and opposition.
• Critics argue that hearings with AI companies focused on investment incentives rather than safety, ignoring existing harms like gender-based violence and discrimination in AI systems.

Background

Australia is positioning itself as a middle power in AI governance, balancing economic growth with security concerns. The country has recently implemented a ban on social media for under-16s and proposed a Digital Duty of Care, signaling a shift toward stricter digital regulation. However, the rapid pace of AI development has outpaced traditional legislative frameworks, prompting calls for principle-based regulations and permanent oversight bodies. The AI Safety Institute, established this year, aims to advise regulators and test safety, but its current resources are seen as insufficient for its broad mandate.

How outlets are covering it

Politico highlights Kate Chaney’s focus on institutional capacity and bioweapon risks, emphasizing the need for sovereign testing capabilities. The Guardian reports on the procedural conflict, with David Pocock accusing the PMO of pre-vetting the committee’s letter, while chairs Briskey and Duniam defend the collaboration as transparent. The Point criticizes the committee’s approach, arguing that hearings with tech firms prioritized investment over safety and ignored existing harms like synthetic abuse content. The Australian was inaccessible due to a bot challenge, so no perspective could be extracted from that source.

Why it matters

The adequacy of Australia’s AI Safety Institute and the integrity of its parliamentary oversight will determine the country’s influence in global AI standards. If the institute remains underfunded, Australia may lack the capacity to enforce safety protocols or protect against bioweapon threats. Additionally, the tension between bipartisan cooperation and perceived government interference could undermine public trust in AI governance. As AI systems increasingly impact labor markets and civil rights, the balance between innovation and regulation will shape Australia’s economic and social stability in the coming decade.

What to watch

The joint-select committee is expected to release its final report by November 30, 2026. Chaney has called for the committee to become a standing body to handle ongoing AI inquiries. The government may consider expanding the AI Safety Institute’s budget and staffing, though no formal decision has been announced. Discussions on mandatory safety testing for synthetic DNA imports may proceed via ministerial action, while debates on copyright and data center levies remain unresolved.