![\"Waleed](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2026\/02\/9e5f0deecf2bfa9b1396af64c30372d0e1efb70f.png\")
Waleed Aly<\/a>Columnist, author and academic<\/p>\nJuly 24, 2026 \u2014 5:00am<\/p>\n
July 24, 2026 \u2014 5:00am<\/p>\n
Save<\/p>\n
You have reached your maximum number of saved items.<\/p>\n
Remove items from your saved list<\/a> to add more.<\/p>\n AAA<\/p>\n The Albanese government is finding its voice on technology. It has taken a curious path to get to this point, and on AI it has been led to it. But having been so led, it now celebrates leading the world. \u201cAustralia will be the first country in the world to bring these issues into a single, national framework,\u201d the prime minister said last week, announcing new national standards for anyone wanting to build AI data centres here. These standards began life in March as environmental \u201cexpectations\u201d, the government curiously refusing to give them the force of law. Now they will be legislated.<\/p>\n This legislation is best seen as one of a pair, its sibling being the government\u2019s teen social media ban. Albanese proudly boasts that \u201cmore than 20 nations have implemented or are implementing social media age restrictions of their own\u201d, following his government\u2019s lead. The success of the policy is, at this point, far from assured, but the government is clearly convinced its message resonates. It symbolises the idea that we have the power to leash the rapacious effects of the technology that has so quickly taken over our lives.<\/p>\n That itself is a watershed notion, a new public mood that is less inclined than previously to assume technology always makes life better. Social media benefited from that assumption, doing all kinds of damage along the way. The truth is, it\u2019s probably too far gone. But \u201cimagine if the world had acted a decade ago\u201d, mused Albanese. \u201cThat is the opportunity \u2013 and the choice \u2013 we now have with artificial intelligence\u201d. If so, it\u2019s an opportunity the government had been choosing to ignore. Until now, it had sung in unison with the Productivity Commission, which hymned about AI\u2019s productivity promises and recommended against any AI-specific regulation. Now the government wants to establish an Office for AI within the prime minister\u2019s own department.<\/p>\n Perhaps the most radical shift, though, is philosophical. Albanese\u2019s speech made a point of trying to see AI not just in economic terms. He began with a rejection of the economics of Thatcher and Reagan, and especially Thatcher\u2019s dictum that there is no such thing as society. He explicitly framed AI as a legal, social, and even moral and spiritual issue, citing the Pope\u2019s AI encyclical along the way. The about-face is dramatic.<\/p>\n The change, I suspect, has three main drivers. In no particular order: the agitation of the Australian Council of Trade Unions ahead of the Labor Party Conference, which has been sounding the alarm about AI-caused job losses; the first signs of community discontent over planned data centres in places like the Blue Mountains; and One Nation\u2019s attack on the government\u2019s original AI policy, which Pauline Hanson described as deciding \u201clargely to leave it in the hands of corporations with little regulation to ensure that the industry operates in the public interest\u201d. All of these criticisms assess AI in terms that are not purely economic. They place questions of quality of life, and not merely wealth creation, at the centre of things. As such, they each capture perfectly a widespread anxiety the government had been avoiding.<\/p>\n That\u2019s what the Albanese government\u2019s new approach is designed to address. The promised data centre standards are aimed mostly at energy and water use. They will require investors to contribute their own renewable power, and that the centres\u2019 demands for resources will not cause energy prices to spike. The government wants Australia to develop its own AI as a way of retaining sovereignty \u2013 of shaping AI rather than being shaped by it. And the overarching promise is that by taking control now, we can ensure AI will deliver jobs, not take them. To that end, Albanese cites labour market research \u2013 \u201cthe first research of its kind by any government anywhere in the world\u201d \u2013 showing graduate employment remains high and jobs in software and tech are growing.<\/p>\n Related Article That\u2019s encouraging news for the moment. But analysis from the US, which is further down the AI path than we are, is less encouraging. In April, Goldman Sachs estimated that AI was causing a net elimination of 16,000 jobs a month. Two months later, it reduced that figure to 11,000, but the difference was mostly down to the jobs being created in building data centres. These jobs are overwhelmingly short term, and once data centres are built, they require few staff to operate. As for the job losses, they\u2019re hitting young people hardest. Goldman Sachs\u2019 data is now starting to show that industries with higher AI adoption rates have higher unemployment among workers younger than 30. It\u2019s an early trend, so it will need to be monitored further. And there will be plenty of opportunity to do so because the rate of AI adoption is only increasing.<\/p>\n And this distils the main tension in the government\u2019s new AI discourse. It talks a more social language, which is welcome. It claims to defend our non-economic goods, like sovereignty and the environment. But it also has to continue to court AI investment, including in the form of data centres, because it sees economic dividends too large to ignore. To resolve this tension, it simply must adopt a sunny view of what government policy can achieve. \u201cWe should not treat AI as a threat to good jobs,\u201d declares Albanese. \u201cWe must use it as an instrument to help create them\u201d, just as we have with technologies past. How, exactly, is not yet known.<\/p>\n The conceit here is that AI is a comparable technology to what has come before, controllable in a comparable way. The trouble is that so many of its creators say the opposite: that even they don\u2019t fully understand how it works or what it will do. We\u2019re yet to see if the government has any answer to that concern. For now, it seems not to share it. Perhaps it sees no alternative, knowing that for all the reassurances of sovereignty and control, AI\u2019s future is mostly being written elsewhere.<\/p>\n Whatever the case and for all the change in the government\u2019s perspective, this optimism that AI remains a mere tool, an extension of the past rather than a departure from it, remains the government\u2019s central, unchanging article of faith.<\/p>\n Waleed Aly is a broadcaster, author, academic and regular columnist.<\/p>\n The Opinion newsletter is a weekly wrap of views that will challenge, champion and inform your own. Sign up here<\/a>.<\/p>\n Save<\/p>\n You have reached your maximum number of saved items.<\/p>\n Remove items from your saved list<\/a> to add more.<\/p>\n![\"Illustration:](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2026\/07\/bb5b788bb667c19254dc20d92f5f00b9718e1061.jpeg\")
Illustration: Simon Letch <\/p>\n![\"](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2026\/07\/38e83a323da2a3cb92881d45169e83344e41a6e1.jpeg\")
<\/a><\/p>\n
![\"Waleed](\"https:\/\/www.newsbeep.com\/au\/wp-content\/uploads\/2026\/04\/9e5f0deecf2bfa9b1396af64c30372d0e1efb70f.png\")