In short:
The federal government wants AI developers to demonstrate their safety systems are effective under proposed laws drawing on banking and aviation regulation.
asserted
systems → want → regulation
It argues voluntary safeguards are inadequate when companies face intense pressure to develop increasingly powerful technology.
asserted
companies → argue → technology
Labor is finalising national AI standards due by the end of the year with plans to legislate the new rules in 2027.
asserted
Labor → finalise → 2027
Australia would require AI companies to prove they are properly testing models and preventing harm under proposed new laws modelled on banking and aviation safety rules.
asserted
they → require → rules
The federal government is finalising plans to regulate frontier artificial intelligence through national standards due by the end of the year, with recent incidents involving rogue agents underscoring the risks of the rapidly advancing technology.
asserted
incidents → finalise → technology
Assistant Minister for Science, Technology and the Digital Economy Andrew Charlton has signalled the government wants to make AI companies demonstrate their safety systems are effective, rather than dictate how they should manage every risk.
asserted
they → signal → risk
He outlined the case for tougher frontier AI oversight in a speech in Sydney today, arguing Australia should put the "burden" where the risk was created.
asserted
risk → outline → burden
"AI needs regulating because its harms are severe, hard to undo, borne by people who never chose them and sometimes invisible until they arrive,"he said.
asserted
they → need → them
"The market will not fix this alone, because the incentives reward speed and capability, and even the people at the top of the industry cannot slow down by themselves."
asserted
people → fix → themselves
Safety rules for banks, aviation show path for AI
Rather than trying to keep pace with the technology by banning specific dangerous behaviours, Australia's approach would require AI companies continually test for risks and show they can prevent harm.
asserted
they → show → harm
Mr Charlton said there was not enough incentive for companies to comply with "voluntary" regulations or codes when it came to frontier AI risks.
asserted
it → say → risks
"You cannot ask firms to volunteer against their own competitive interest amid a manic race involving trillions of dollars and then be surprised when they do not," he said.
asserted
he → ask → dollars
He also warned detailed prescriptive rules could end up "out of date before the ink is dry" due to the pace of the technology's development.
uncertain
ink → warn → development
"For these reasons, there is much that Australia can learn from systems-based regulatory approaches, that puts the onus and accountability on companies for managing how they develop products and services safely," he said.
asserted
he → be → products
Mr Charlton said this model was already used in workplace health and safety, supervision of banks and aspects of aviation safety.
asserted
model → say → safety
"The goal is to ensure companies are meeting the expectations of safety, without trying to specify every hazard," he said.
asserted
he → ensure → hazard
"The question isn't only 'did something go wrong' but is there a serious system in place to detect risks and prevent incidents occurring.
asserted
incidents → go → risks
"
AI race not one Australia can 'win' or 'sit out'
asserted
Australia → win → ?
The push for tougher regulation has been sharpened by revelations OpenAI agents gained unauthorised access to Australian government websites, including a Medicare statistics portal, during testing.
asserted
agents → sharpen → testing
OpenAI has now backed mandatory safety requirements, independent assessments and incident reporting, while also calling for international standards.
asserted
OpenAI → back → standards
The company's chief executive, Sam Altman, told the United Nations last month the "most important decisions cannot be made by labs in San Francisco alone".
asserted
decisions → tell → Francisco
"We need accurate and speedy incident reporting, classification and reporting protocols, so the world can learn from failures before they become catastrophes," he said.
asserted
he → need → failures
Mr Charlton said frontier AI labs were currently putting "capability ahead of safety" as they raced each other, with a broader competition also playing out between the US and China.
asserted
competition → say → US
He said Australia's task was not to "win" that race or "sit it out" but to "change it so that this technology is built to serve people".
asserted
technology → say → people
Australia must 'host and build' AI
Mr Charlton said safety standards alone would not be enough due to "hostile states or criminals who were never going to comply".
asserted
who → host → states
"Relying on regulation alone is like building a fence around a horse, only to realise that horse has wings," he said.
asserted
he → rely → wings
Mr Charlton said if Australia wanted a say in how AI technologies were built and governed it needed to be a country that could "host and build them".
"This is our chance to shape this technology, for our standards to set the rules of the road globally over decades," he said.
uncertain
he → say → decades
Shadow Foreign Minister Ted O'Brien has called for Australia to strike a deal with the United States to secure preferential access to the most advanced AI models and strengthen defences against cyber attacks.
asserted
Australia → call → attacks
"We can use AI as a shield to counter AI being used as a sword,"he said.
asserted
We → use → sword,"he
In a speech at the University of Technology on Wednesday, Mr O'Brien said Australia should pursue a government-to-government agreement guaranteeing continued access to American technology, while developing its own sovereign AI capabilities.
asserted
Australia → say → capabilities
The Aporia
