It took 20 years to start regulating social media - will AI controls come sooner?

Analysis: Australian Prime Minister Anthony Albanese's decision to reveal a three-month old hack of his country's Medicare portal by OpenAI was a calculated political flex.Albanese is under pressure at home, sagging in the polls, and struggling with a series of gaffes and policy U-turns. In New York for the UN General Assembly, this was a chance to burnish his record for straight-talking on the tech-bros.But there was a deeper, much more globally significant side to the announcement.Why the latest OpenAI 'hack' disclosure really mattersCanberra has been quick to point out that the OpenAI hack itself was relatively small beer. There is nothing to suggest patient data was accessed, and the errant agent seems to have got no deeper into the sprawling Services Australia network.This perhaps explains why the notification OpenAI provided Australia was - according to Albanese - tardy and "unacceptable" in its delivery.Choosing to disclose the hack publicly within hours of shirt-fronting OpenAI chief executive Sam Altman about it was, therefore, clearly calculated by Albanese for maximum impact.Albanese has form for tackling big tech. His government was the first to implement a ban on social media for young people. It has just passed legislation forcing Meta, Google, TikTok and LinkedIn to strike deals with local publishers or face a levy. And it's now on to a 'digital duty of care' bill that proposes forcing social platforms to take more responsibility for the harms they cause.All of Australia's actions have been fiercely criticised by the tech companies. In the Trump Administration, those companies now have even more powerful back-up.It is rare for one nation to speak up on another's domestic policy but the US Embassy in Canberra this week made a public submission against the 'digital duty of care' bill, saying the US would likely see such a measure as "facilitating censorship".Anthony Albanese has been willing to take on America's powerful big tech companiesAFP / David GrayHow the mood for AI regulation differs from social mediaOpening up a new front on AI in this context will be weighed up by Albanese against the ever-present threat of Trump reacting angrily in ways that could really hurt his nation's interests.But on AI, Australia is less isolated than on social media. It is dealing in technology that is many times more consequential for humanity than social media - which itself has been hugely impactful.This week, Australia was one of 22 so-called 'middle powers' that signed on to a statement arguing for global controls of AI. They were effectively echoing statements from the leaders of the frontier AI labs, including OpenAI's Altman, calling for the same.Trump has pushed back against "any attempt to construct a globalist scheme to control artificial intelligence". But even he has admitted there is a need for care and there are reportedly voices within his administration gaining ground in the case for control. US Treasury Secretary Scott Bessent, for one, has put some time in with China's vice-premier in search of a notification system on AI safety.Actual regulation does not yet exist. Early efforts appear to have focused on the physical infrastructure - data centres - but the public discourse has been largely filled up with declarations of concern and generic calls for urgent action.There seems now to be widespread agreement on the need for action, which took many years to arrive for social media. In fact, the public appetite for AI regulation has now very likely raced ahead of that for social media.U.S. Treasury Secretary Scott Bessent met with Chinese Vice Premier He Lifeng in New York at the weekend.Xinhua via AFPHands up if you have a planCould the levers of state and international diplomacy move as quickly as might be required to make AI regulation a reality? If the norm-smashing Trump Administration has shown the world anything, it's that the usual processes for rule-making can be bypassed if there is a strong enough will.The next step, then, for leaders like Altman, Bessent and even Albanese is to start describing the actual shape AI regulation could take.There is really no reason why it couldn't be someone else entirely, however, who comes up with the safety plan the world seems to want for AI right now.We've arrived at a point where anyone with the outlines of a workable plan should raise their hand and speak.On the list of 'middle powers' that signed on to the call for AI guardrails were Norway, Finland and Singapore - all nations with similarly-sized populations to New Zealand. Anyone with the smarts, charisma, and connections from these or really any other nation could emerge to put this impulse into flight.Why not, for instance, the co-founder of Google DeepMind, Shane Legg, who has been heavily involved in AI safety? He went to Rotorua Lakes High School and studied at Waikato University. Maybe it will be him?