Opinion
The intelligence may be artificial, but the reactions to it are very real.
New York State this week announced a 12-month ban on the building of big new data centres.
The world was in the throes of an economic upheaval on a scale unseen in generations – “perhaps ever”, said the state’s governor, Kathy Hochul.
“These hyperscale AI data centres consume enormous amounts of power, truly threatening to outpace our grid’s capacity. They drive up costs for local ratepayers, and I refuse to let those costs get passed down to New Yorkers,” she said.
In the same week, some 40,000 workers at carmaker Hyundai struck back at company plans to introduce an intimidatingly large humanoid robot into its factories.
Hyundai’s “Atlas” walks and moves freely and can lift 50 kilograms. South Korean president Lee Jae Myung says of AI-driven robotics that it’s “impossible to avoid the giant chariot that is rolling on”.
Perhaps, but workers at Hyundai, the world’s third-biggest carmaker, have committed to only work half days until bosses guarantee job and wage protections against the android, which stands at 190 centimetres and weighs 90 kilograms. It’s a global test case.
And a historical first in war occurred when Iran launched missiles at data centres in two US-allied Gulf nations this year. It was the day after the US-Israeli attack began.
It “marks a sea change in warfare”, says the Centre for Strategic and International Studies’ Emily Harding, and it is the first known military attack on data centres anywhere.
The reason? Iran said it attacked the Amazon Web Services facilities “in order to identify the role of these centres in supporting the enemy’s military and intelligence activities”.
And so “the strikes swiftly brought the war directly into the lives of 11 million people in the UAE,” as The Guardian reported. Emiratis learnt the hard way that neither the US nor their own government were able to protect them.
“Millions of people in Dubai and Abu Dhabi woke up on Monday unable to pay for a taxi, order a food delivery, or check their bank balance on their mobile apps,” as The Guardian said.
Australians affected by the recent Telstra failure would recognise the symptoms. One difference is that while Telstra restored services within a day or two, the missiles caused physical damage to the UAE that caused persistent interruptions for weeks. Iran also attacked a third data centre, in Bahrain.
“The disruption from the Gulf strikes is a prelude to future conflict,” Miah Hammond-Errey writes for the Lowy Institute.
These three AI-related events – a moratorium on new data centres in New York, a strike over humanoid robots in South Korean factories, and the first known military attack on data centres – illuminate the problems being wrought by AI, but also the priority status it holds.
“It’s like water,” says a former chief of Australia’s highly sophisticated cyber spy agency, the Australian Signals Directorate, Rachel Noble. “It’s flowing. It’s happening, and we’re all using it,” she tells me. It’s quickly become a fundamental input to modern societies, but one that we take for granted.
“It’s coming, and it’s here. So I think we also have to understand that reality, and not a lot’s really going to stop it.” Beyond consumer fun and frolics with chatbots, web searches and companion bots, “we’re already highly dependent on artificial intelligence, for example, on space with GPS, or cybersecurity”.
An enemy state will try to demolish a country’s AI capability because “it’s the modern equivalent of World War II bombing the dam. That inhibits the water supply. So people without water, it’s getting at the very heart of what potentially sits behind the normal operations of your civil society.”
Iran did so with missiles in two Gulf nations. China, as the Australian Signals Directorate has reported, has set time bombs to do the same digitally, by planting malicious pieces of software in Australian critical infrastructure.
“Some of the most concerning recent advisories are telling us that the People’s Republic of China has the capability, and has been found in Australia pre-positioning on our critical infrastructure,” Noble tells me on the sidelines of the annual Australian-American Leadership Dialogue in Washington this week.
“It’s extremely hard to find because it’s not active, and it’s sitting there so that that malicious software can be activated at a time of their choosing, and that means that they would be able to deny, degrade, disrupt services provided through that critical infrastructure, or completely destroy it. That helps us understand what the PRC’s intent might be towards Australia.” AI is crucial to detect and deflect cyberattacks, she says.
Prime Minister Anthony Albanese this week issued Australia’s first high-level attempt to channel the water, to set out the principles that Australia will impose on AI, to keep the water flowing, to seize the opportunity, but to save us from drowning. Or, conversely, dying of thirst imposed by foreign malice.
The prime minister’s speech was an important marker for the world. It set out a framework while other countries, including the US, are still fumbling to find the best approach to this onrushing revolution.
Australia has a moment in time to manage some of the biggest problems that beset the US and others because data centre investment started more slowly. Australia, for instance, will insist that data centres add more to the electricity grid than they draw. And that workers’ rights will be protected, not trampled.
Albanese’s foundational speech was timely for one obvious reason, and for a second, unpublicised reason that’s altogether more important.
The obvious reason is that he was pre-empting Labor’s National Conference next week. The rank and file would like to put onerous regulation on AI. Albanese signalled to his party that he will legislate for AI in Australia, but flexibly, to attract investment, he will not deter it.
“We are serious about attracting frontier AI investment to Australia. Because we want AI to support and create good jobs, not replace them,” he said. “It is not our goal to try and legislate for every possible eventuality or risk. That only creates the risk of Australia missing out on investment altogether. This is about having the flexibility to keep pace with change, and get out in front of it.”
The unpublicised reason is that, behind the scenes, the Albanese government has been in intense negotiations with the White House on AI.
Specifically, the government is seeking preferential treatment for Australia to get first-release access to the newest frontier models of AI. The Trump administration had earlier decided on a laissez-faire approach, allowing AI producers to release their latest offerings as they pleased.
But the White House abruptly changed policy on June 12. It instructed Anthropic to suspend access to its potent Mythos 5 and Fable 5 models for any foreign nationals.
The administration cited unspecified national security concerns. In fact, the fear was that hostile governments could use them to find heretofore unknown vulnerabilities in America’s own cyber shield and force their way into its most sensitive intelligence and military systems. This export control order was lifted on June 30.
But the White House is still wrestling with the question of consistent standards and rules it should apply for foreign access to cutting-edge models. Donald Trump was expected to issue an executive order last week, so a ruling is imminent. Australia’s hope is that it will be favoured as a trusted ally and a Five Eyes partner. Albanese explicitly cited the Five Eyes alliance in his speech.
Noble explains the stakes: “Within the Five Eyes intelligence agencies, it’s their primary job to provide support to our war fighters. People feel uncomfortable that we talk in binary terms. But if there is war, we must win. And why on earth would we diminish our chances by not adopting the world’s best and most powerful technologies? It’s really important that we don’t let ourselves fall behind.” China is America’s only peer competitor in AI.
Albanese’s speech won a rare double plaudit, attracting the approval of both the Business Council of Australia and the ACTU. The BCA was pleased to see that Albanese wanted to enable investment; the ACTU that he wanted to protect the rights of workers and the copyright of creative artists.
Labor senator Deborah O’Neill is in Washington this week in her capacity as the inaugural chair of the Australian parliament’s new defence committee. She reports that, in her meetings with US officials and legislators, “there were absolutely no concerns expressed” about Albanese’s statement and that “the feedback has been very, very positive”.
We’ll soon see if the government succeeds. For her part, Noble was pleased to see Albanese taking leadership on AI, but she says: “I think the homework will be tested about just how fast they can move because I think we don’t have years to contemplate this challenge. We have weeks, months.”
The intelligence may be artificial, but the competitive urgency is real.
Peter Hartcher is political editor and international editor. He writes his world column on Tuesdays.
Get a note directly from our foreign correspondents on what’s making headlines around the world. Sign up for our weekly What in the World newsletter.
Peter Hartcher is political editor and international editor of The Sydney Morning Herald and The Age.Connect via email.From our partners
Discover more from PressNewsAgency
Subscribe to get the latest posts sent to your email.