The AI Safety Institute will support model testing, regulatory responses and safe AI governance in Australia’s interests.
Australia’s AI Safety Institute became operational on 2 June as the government seeks to strengthen public trust in AI development, deployment and governance. The announcement was made during the AFR AI Summit in Canberra, where the government described public trust as essential to building a domestic AI industry.
According to Assistant Minister for Science, Technology and the Digital Economy Hon Dr Andrew Charlton, Australia’s national AI plan rests on three pillars:
- Capturing the opportunity
- Sharing the benefits
- Keeping Australians safe.
The AI Safety Institute is intended to support that effort by testing AI systems, assisting regulators and strengthening public confidence in the technology.
In his speech, Charlton also argued that Australia faces a choice between building a world-class AI industry or relying on foreign capability, while warning that low public trust could slow AI adoption and investment.
Charlton cited survey findings showing that only 30% of Australians believe the benefits of AI outweigh the risks, while 78% are concerned about potential negative impacts, and 36% say they trust the technology. It linked public scepticism to concerns that AI benefits may flow offshore while costs linked to jobs, privacy, power bills, and local communities are borne domestically.
Data centres were highlighted as an example of how trust considerations are shaping AI policy. The government said data-centre developers should contribute new renewable energy capacity, cover an appropriate share of transmission and distribution costs, engage with local communities and avoid creating pressure on water resources.
The AI Safety Institute will analyse and test AI models and applications, support regulators responding to emerging AI-related harms, and contribute to national and international discussions on safe AI development and governance. The speech also pointed to wider work on privacy reform, online safety, workplace impacts, competition, consumer issues, and public-sector AI adoption.
Why does it matter?
Australia is positioning trust as a key component of its AI strategy at a time when governments are balancing economic opportunities from AI with concerns about safety, privacy, employment and infrastructure impacts.
By creating a dedicated AI Safety Institute, Australia joins a growing number of countries establishing specialised institutions to evaluate AI risks, support regulators and build public confidence in the deployment of increasingly capable AI systems.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!