AI agency a ‘toothless tiger’ without sovereign capability
The government has unveiled its AI road map. An expert is calling for action on a key element, saying it could wind up a ‘toothless tiger’.
The newly announced artificial intelligence research agency risks “becoming a toothless tiger” if Australia continues relying on foreign AI models, an expert championing sovereign capability has warned.
The Albanese government on Tuesday announced it would pump nearly $30m into setting up the AI Safety Institute while unveiling its National AI Plan.
The new agency, flagged last week, will be ready to go in early 2026 and focus on key parts of the plan’s goals.
Simon Kriss is a leading AI strategist and the chief executive at Sovereign Australia AI – the firm building the country’s first large language model.
He welcomed the government’s road map but said Australia must ditch its dependency on “on foreign AI powers”.
“The Australian government has landed in the right place with sweeping regulation of AI,” Mr Kriss said.
“However, this does not remove the clear and present danger of relying on foreign-made AI models.
“The announced AI Safety Institute is at risk of becoming a toothless tiger if all our AI is purchased from overseas where they care less about our values and laws.
“For Australian businesses to begin to trust in and adopt AI, we must be assured that the models we use are built under Australian law and that none of our data ever leaves Australian shores or is processed by servers owned by American companies who are subject to the US CLOUD Act.”
The US’ Clarifying Lawful Overseas Use of Data (CLOUD) Act lets US federal law enforcement force tech firms to hand over requested data regardless of where it is from.
Developing sovereign AI capability is a key to the national plan
As part of guarding against malicious actors, Australia will develop GovAI as a “centralised AI hosting service, providing a secure, Australian-based platform for agencies to develop customised AI solutions at low cost”.
“Locally developed models may provide benefits such as reflecting unique cultural context or language, supporting innovation and mitigating risks around data security,” the plan says.
“The government is backing local capability through significant investment in sovereign AI for the public service.
Another part of its “proactive” approach is leading international discussions around rules.
“Shaping global governance of AI is vital for Australia’s economic prosperity and national security,” the plan says.
“Australia can use its role as a responsible middle-power to embed our values of safety, transparency and inclusion in international AI norms and standards.”
‘New and unknown threats’
Industry and Innovation Minister Tim Ayres said the National AI Plan “is about making sure technology serves Australians, not the other way around”.
“This plan is focused on capturing the economic opportunities of AI, sharing the benefits broadly, and keeping Australians safe as technology evolves,” Senator Ayres said.
He went on to say that AI would “help close gaps in essential services, improve education and employment outcomes, and create well paid jobs in future industries”.
“Guided by the plan, the government will ensure that AI delivers real and tangible benefits for all Australians,” he said.
“As the technology continues to evolve, we will continue to refine and strengthen this plan to seize new opportunities and act decisively to keep Australians safe.”
While big tech and governments alike promise AI will unleash a new era of economic growth, national security experts are sounding the alarm on its potential weaponisation, including by state-backed actors.
The plan itself warns AI will supercharge national security challenges already facing the country and “create new and unknown threats”.
It also says the Department of Home Affairs and law enforcement agencies will “proactively mitigate the most serious risks posed by AI”.
Keeping Australians safe also means recognising that AI is likely to exacerbate existing national security risks and create new and unknown threats,” the plan says.
“To keep Australians safe, the government is taking proactive steps to prepare for any potential AI-related incident.”
Originally published as AI agency a ‘toothless tiger’ without sovereign capability
‘30 choppers’: Chilling details, ADF watching
