Treasurer Jim Chalmers has thrown his weight behind artificial intelligence as a fix for Australia’s stubborn productivity problem, arguing that a country which learns to use the technology well will be better placed to grow wages, ease cost-of-living pressure and, eventually, bring interest rates down.
In comments reported by the ABC, Chalmers drew on a meeting he held with United States AI developer Anthropic in April, using it to sketch out why he believes the technology is now a macroeconomic issue rather than just a corporate IT one. You can read the ABC’s account of his remarks here.
Why productivity became the Treasurer’s obsession
To understand why Chalmers is reaching for AI, you have to understand the hole he is trying to climb out of. Productivity, the amount of output an economy squeezes from each hour worked, is the single biggest driver of living standards over the long run. When it grows, businesses can pay more without lifting prices, governments collect more revenue without raising taxes and households get ahead. When it stalls, everything gets harder.
And in Australia it has been stalling for a long time. Productivity growth has been slowing for the better part of two decades, and the past few years have been particularly weak. The Reserve Bank and Treasury have both flagged that without a turnaround, the pay rises Australians have come to expect simply are not sustainable. That is the backdrop against which the Treasurer is making his pitch: the old levers are not delivering, so something new has to.
The link Chalmers is drawing to interest rates is the part that will grab attention. His logic runs like this: if AI helps the economy produce more without pushing up costs, it takes some of the inflationary heat out of growth. Lower inflation pressure gives the Reserve Bank more room to keep rates lower than they otherwise would be. It is a long causal chain, and one the Treasurer does not control, but it explains why he is treating AI adoption as a national economic priority rather than a niche concern.
The Anthropic meeting
The April sit-down with Anthropic, the San Francisco company behind the Claude family of models, appears to have sharpened the Treasurer’s thinking. Anthropic has been unusually visible in Australia this year, with reported plans for large-scale local data centre capacity and a series of partnerships with Australian firms. For a government trying to work out how to ride the AI wave rather than be flattened by it, a direct line to one of the frontier labs is useful intelligence.
It also signals something about the government’s posture. Rather than positioning AI purely as a risk to be regulated, Chalmers is presenting it as an opportunity to be captured, provided Australia moves quickly enough and spreads the benefits widely. That framing sits alongside the government’s broader work on a national AI approach, including the recently announced plans for a dedicated office to coordinate policy across the public service.
Two ways to read the pitch
Supporters of the Treasurer’s argument point to a growing body of evidence that AI can lift output in knowledge-heavy work: drafting, coding, analysis, customer service and the like. If even a slice of the Australian workforce becomes measurably more productive, the aggregate effect on the economy could be significant. On this reading, the government’s job is to remove barriers to adoption, invest in skills and make sure the digital infrastructure is there.
Sceptics are less convinced, and they are not hard to find. Australia has a patchy record of turning promising technologies into measured productivity gains, and there is a live debate about whether the current AI boom is as transformative as its boosters claim. Recent commentary has questioned whether the headline numbers around Australia’s AI investment surge really reflect broad-based productivity, or whether they are concentrated in a handful of data centre and infrastructure plays that do not touch most workers.
There is also the uncomfortable question of who benefits. If AI gains flow mostly to capital and to the largest firms best placed to deploy it, the productivity dividend may not show up in worker pay packets at all. Unions and some economists have warned that without deliberate policy, the technology could widen rather than narrow gaps. Separate departmental analysis this year found AI was not, at least so far, driving Australian job losses, but the distribution of any upside remains an open question.
What it means for Australia
For Australian businesses and workers, the Treasurer’s framing matters because it shapes where policy attention and money will go. If AI is treated as central to the productivity agenda, expect more emphasis on adoption support for small and medium firms, skills programs, and the data centre and energy infrastructure that AI workloads demand. That last point is not trivial: Australia is already wrestling with the strain large data centres place on the electricity grid, and any serious AI build-out will collide with the energy transition.
The interest rate angle is what will resonate most with households still feeling the squeeze of the past few years. It is worth being clear-eyed, though. Even in the most optimistic scenario, AI-driven productivity gains take years to show up in the national accounts, and rate settings depend on far more than one technology. The Treasurer is describing a direction of travel, not a promise about the next Reserve Bank meeting.
There is a sovereignty dimension too. If Australia’s productivity story is going to lean on frontier AI models built offshore, questions about local capability, data governance and who controls the underlying infrastructure become economic questions, not just security ones. That tension, between wanting the productivity upside and not wanting to be wholly dependent on a few foreign labs, is likely to run through the policy debate for years.
What’s next
The immediate test is whether the rhetoric turns into detail. A speech about AI and productivity is easy; a coherent plan covering adoption incentives, skills, infrastructure and worker protections is harder. Watch for how the government’s emerging AI coordination machinery translates the Treasurer’s ambition into concrete measures, and whether the Productivity Commission and Treasury put numbers behind the claim that AI can meaningfully shift the national dial.
For now, Chalmers has done something notable simply by elevating AI from a technology story to an economic one, and by tying it to the two things voters care about most: wages and rates. Whether the technology delivers on that framing is the question that will define the debate.
Sources: ABC News, Technology.
