Sarah Hunter, the Reserve Bank of Australia’s Assistant Governor (Economic), told a UBS conference in Sydney on 9 November that AI is lifting demand before it lifts supply. In her speech, she said “the positive shock to aggregate demand set against the economy’s current supply capacity will put some upward pressure on inflation”, while there is “limited evidence that AI has materially lifted aggregate productivity”.
An RBA staff note released the same day puts a number on the demand: a further A$115 billion in real terms invested in Australian data centres between mid-2026 and the end of 2028. The staff assume about 70 per cent of data centre investment is imported. Applied to the A$115 billion, that leaves roughly A$34.5 billion of domestic content, by our arithmetic. Spread over the 2.5 years, the total is about A$46 billion of spending a year.
Business Pill · Financial Conditions Explained
An evergreen Business Engineer explainer on financial conditions, how the price and availability of money move markets. It teaches the general idea only and does not discuss the RBA or this speech.
The key insight: The RBA’s Sarah Hunter said AI is adding to demand before it adds to supply: in the near term it will put some upward pressure on inflation, through pricier consumer electronics and construction costs, while there is limited evidence it has lifted productivity. RBA staff expect a further A$115 billion of data centre investment by the end of 2028; with about 70 per cent assumed imported, roughly A$34.5 billion is domestic content, by our arithmetic.
The speech says surging global AI investment has contributed to capacity pressures in supply chains for products such as semiconductors, and that “Price increases have been concentrated in a narrow set of consumer electronics goods that have small weights in CPI baskets”. It adds that “some liaison contacts are reporting that data centre construction is adding to demand for some already-stretched construction skilled trades”.
In the question-and-answer session, Hunter named the goods: “game consoles, the latest mobile phones, and that sort of thing. We’re seeing the prices of those rise fairly sharply now”. She described a second channel, construction costs, which in the CPI is “captured through the new dwelling construction cost component”.
She also limited the role of AI in the RBA’s latest inflation forecasts. A substantive part of the upward revision, “particularly for headline, is due to reflecting the oil price profile that we’re now assuming”, she said, and “AI and data centres are part of it, but they’re not the whole story”. Over time, the speech says, “AI’s long-run effect on inflation is likely to be broadly neutral”.
We measured the direct channel on 3 November, the day the RBA raised its cash rate to 4.85 per cent: ABS data showed the index for audio, visual and computing equipment rose 6.1 per cent in the September quarter, and the RBA’s own “AI-related goods” category is 1.6 per cent of the CPI basket.
RBA staff expect a further A$115 billion in real terms to be invested in Australian data centres between mid-2026 and the end of 2028, excluding supporting infrastructure, and assume around 70 per cent of data centre investment is imported (RBA Insight 2026-11, 9 November 2026). The A$80.5 billion and A$34.5 billion split and the A$46 billion a year are our arithmetic.
What Was Posted on X
Bloomberg: Reserve Bank of Australia Assistant Governor Sarah Hunter says Australiaโs AI investment surge will lift inflation before productivity gains arrive.
Data centre construction is adding demand for already-stretched skilled trades, and higher building costs are reaching the consumer price index.
Pricier new phones and game consoles add a smaller push, she said.
AI has barely touched headcount, but job ads and employment are weaker in AI-exposed occupations, especially for younger workers.
She warned equities and corporate debt could reprice sharply if AI returns disappoint, since investors demand little compensation for risk.
Rohan Paul’s post on 9 November relayed a Bloomberg report on Sarah Hunter’s speech, with a screenshot of the article. We checked each point against the RBA’s own speech text, its Q&A transcript and the staff note on data centre investment, and quote those documents above.
The A$115 Billion Pipeline
The staff note, Data Centre Investment in Australia, says investment has grown around 10-fold over the past three years. Because much of the equipment is imported, data centres “have directly accounted for up to around 0.2 percentage points of annual GDP growth over the past year”. The building is about a quarter of the cost and the servers, storage and networking equipment about three-quarters.
The A$115 billion forecast is “equivalent to almost 2 per cent of GDP each year in gross terms”, the note says, and excludes supporting infrastructure. Box A compares it with the mining boom: “comparable to the early stages of the mining boom but well below its peak”.
On labour, the note cites SEEK data showing demand for roles supporting data centres “has more than doubled since mid-2025”. Contacts say data centre projects are “a top bidder” for electricians, plumbers, HVAC technicians and site supervisors, and that “In a supply-constrained economy, this could place additional upward pressure on construction costs and wages”.
Jobs: Little Headcount Effect So Far
Hunter said contacts in the RBA’s liaison program indicate “that AI has not yet impacted most firms’ headcount” and is not expected to over the next year. The speech points to “tentative signs of adjustment, through weaker job ad posting” and employment outcomes in more AI-exposed occupations, “particularly for younger workers”, but adds that “these differences are modest”.
Hunter said the exposure of financial market participants to the AI boom is rising “at a time when risk premia in global financial markets are low by historical standards”. Her warning: “If the returns to AI investment disappoint, some asset prices (including equities and corporate debt) will be exposed to sharp repricing and possibly disorderly adjustments.”
Staff estimate that “around 5 per cent of Australian households’ financial wealth is invested in the equity of AI-related companies”. She also said the RBA’s estimates of the neutral rate have risen, with market participants citing strong demand for capital as a driver of higher bond yields. In the Q&A, she said the RBA’s productivity assumption through its forecast horizon “is 0.7 per cent year on year”, adding: “That’s the assumption that we changed to last August, and we haven’t changed that since.”
As we read it, the speech splits AI’s price effect into a small, visible channel (phones and consoles, a small weight in the CPI) and a broader, slower one that runs through construction trades and new dwelling costs.
As we read it, the import share is what keeps the A$115 billion from being a mining-boom-sized shock to domestic demand: most of the money buys equipment made in Asia, while the part spent in Australia competes for the same electricians and site supervisors as other building work.
Sarah Hunter, RBA Assistant Governor (Economic), 9 November 2026
“If the returns to AI investment disappoint, some asset prices (including equities and corporate debt) will be exposed to sharp repricing and possibly disorderly adjustments.”
Three Implications
FOR INFLATION WATCHERS Hunter said the direct channel is in goods with a small weight in the CPI; the second channel, construction costs, shows up in the new dwelling construction cost component.
FOR BUILDERS AND TRADES RBA staff say data centre projects have become a top bidder for transferable trades, and SEEK data show demand for roles supporting data centres has more than doubled since mid-2025.
FOR INVESTORS Staff estimate around 5 per cent of Australian households’ financial wealth is in the equity of AI-related companies, while Hunter said risk premia are low by historical standards.
The framework describes the AI build as having “a dual character that is simultaneously a bubble and a supercycle”.
As we read it, Hunter’s speech holds both sides: a demand boom and a possible repricing of equities and corporate debt now, productivity gains later. The framework predates the speech and does not discuss it.
What Is Not Established
The speech does not quantify how much AI adds to Australian CPI. Hunter said much of it is “hypothetical” and that “It’s too early to say how large the effects will be or predict precisely when they will emerge”. The A$115 billion is a staff forecast, and the note carries a disclaimer that its views are the authors’, not the RBA’s. Hunter called the risks to the investment outlook balanced, with “very wide error bands”.
The 70 per cent import share is an assumption within external estimates of 50 to 80 per cent, so our domestic split is only as good as that assumption.
This Week In The AI Supercycle
The Great Convergence of Capital, Power, and Geopolitics
This week’s issue: for two years the AI build-out was priced as a supply story, and last week as a credit story. This week it was priced as a permission story, and the answer arrived as a cost, not a ban.
The RBA’s Sarah Hunter said AI is a demand shock first: it is adding some upward pressure on inflation through consumer electronics and construction costs, while productivity gains have yet to show up. RBA staff expect a further A$115 billion of data centre investment by the end of 2028. With about 70 per cent assumed imported, roughly A$34.5 billion of that would be domestic content, by our arithmetic.
A note on sourcing. Quotations are from the RBA’s published text of Sarah Hunter’s 9 November 2026 speech and its question-and-answer transcript. The data centre forecast and import assumption are from RBA Insight 2026-11, whose views are the staff authors’. We found the story through an X post relaying Bloomberg and did not read the Bloomberg article. The A$34.5 billion and A$46 billion figures are our arithmetic.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
