Anthony Albanese has signalled that reforms to ensure \u201cfairness\u201d is embedded in the rollout of artificial intelligence in Australia are next on his reform agenda.<\/p>\n
After securing Senate support for tax reform to overhaul capital gains tax and negative gearing, the Prime Minister has turned his attention to AI.<\/p>\n
His speech follows the concerns of international spy agencies about cybersecurity, that rapid developments in AI models could see the world face faster cyber attacks that are more complex and difficult to contain.<\/p>\n
In a speech to the Centre for Economic Development in Australia, Mr Albanese said it was important that \u201cnational values\u201d informed the rapid expansion of artificial intelligence, but was light on detail about what it all means.<\/p>\n
\u201cWe can bring our national values of fairness and opportunity to AI, so that it grows our economy and strengthens our sovereignty, without fragmenting our society or damaging our environment,\u201d he said.<\/p>\n
\u201cAnd we can own our place in the architecture of the fastest region of the world in human history.\u201d<\/p>\n
Greens Senator Sarah Hanson-Young has warned against a \u201ctech bro free-for-all\u201d that could drain Australia\u2019s power and water resources.<\/p>\n
\u201cWe are sleepwalking into an AI crisis,\u2019\u2019 she said.<\/p>\n
\u201cUntil Australia\u2019s laws are up to the task of regulating the big tech bros and their power-hungry datacentres, they should not be given the green light to drain our power and water.<\/p>\n
\u201cUntil the necessary safeguards are in place and Australia\u2019s resources are protected, approval and development must be paused.\u201d<\/p>\n
Meanwhile, independent MP David Pocock claims that Anthony Albanese is preparing to announce a new artificial intelligence reform agenda next month, potentially offering expedited approvals for new data centres.<\/p>\n
US President Donald Trump\u2019s intervention<\/p>\n
In June, the US Commerce Department used export-control powers to demand that Anthropic restrict the access of foreign nationals to Fable 5 and Mythos 5 models, citing national security concerns.<\/p>\n
Anthropic then pulled both of its newest frontier AI models offline worldwide.<\/p>\n
Citing national security fears, the US issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.<\/p>\n
\u201cThe net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance. Access to all other Anthropic models will not be affected,\u2019\u2019 Anthropic said in a statement.<\/p>\n
\u201cOur understanding is that the government believes it has become aware of a method for bypassing, or \u201cjailbreaking,\u201d Fable 5.<\/p>\n
\u201cFable\u2019s safeguards are substantially more effective than those of any previously deployed model. No testers have yet been able to find a universal jailbreak \u2013 a jailbreak method that can very broadly bypass the model\u2019s safeguards, unblocking a wide range of cyber capabilities.<\/p>\n
A jailbreak refers to the process of manipulating the AI into bypassing its built-in safety guardrails so it can generate restricted or high-risk outputs. <\/p>\n
What are the risks of Fable 5<\/p>\n
The Fable 5 controversy reflects the US government\u2019s fears that the model\u2019s high-level capabilities in cybersecurity and coding flaws could be weaponised by bad actors.<\/p>\n
\u201cWe are complying with the government\u2019s legal directive and are removing access to Fable 5 and Mythos 5 for all users,\u201d the statement read.<\/p>\n
\u201cHowever, we disagree that the finding of a narrow potential jailbreak should be cause for recalling a commercial model deployed to hundreds of millions of people. If this standard was applied across the industry, we believe it would essentially halt all new model deployments for all frontier model providers.<\/p>\n
\u201cAs we have stated publicly, we believe the government should have the ability to block unsafe deployments, as part of a statutory process that is transparent, fair, clear, and grounded in technical facts. This action does not adhere to those principles.<\/p>\n
\u201cWe apologise for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible.\u201d<\/p>\n
Five Eyes warning<\/p>\n
Mr Albanese\u2019s speech comes just days after intelligence agencies for the Five Eyes warned in a rare joint statement that political leaders needed to \u201cact now\u201d.<\/p>\n
The Five Eyes (FVEY) is an intelligence alliance comprising Australia, Canada, New Zealand, the United Kingdom, and the United States.<\/p>\n
\u201cFrontier AI models are anticipated to exceed current industry expectations, fundamentally transforming both offensive and defensive cyber capabilities. The timeline is not years, it is months,\u201d the statement said.<\/p>\n
\u201cIn this environment, cyber resilience is integral to advancing business continuity, market confidence, and long-term value.\u201d<\/p>\n
The cybersecurity agencies had raised concerns that rapid developments in AI models suggested the world could face faster, more complex, and more difficult-to-contain attacks.<\/p>\n
\u201cA whole-of-organisation and whole-of-society response is required,\u201d the statement said.<\/p>\n
\u201cCyber risk can no longer be treated as a purely technical issue. This is a core business risk and leadership responsibility.\u201d<\/p>\n