Artificial intelligence has evolved over recent years to now be the largest driver of capital spending in global equity markets. This paper explains what artificial intelligence is and how it works, examines the unprecedented scale of investment now taking place, and outlines how Vantage portfolios are positioned for the impact AI will have.
What Is Artificial Intelligence?
An artificial intelligence model is a computer program that has been shown enormous volumes of data including text, images and code, and has learned the patterns within that data well enough to produce new work of its own. Ask it a question and it assembles a response one word at a time, choosing at each step the continuation that best fits everything it has learnt. Over time, these models have continuously improved as more data and computing power have increased.
Although the term artificial intelligence was invented by computer scientists back in the 1950s, it was not until OpenAI released ChatGPT to the public in November 2022 that the technology became widely recognised. ChatGPT reached an estimated 100 million users inside two months, faster than any consumer product before it. Since then, many different artificial intelligence models have been released; Gemini, Grok, Bard, including Anthropic’s Claude, naming but a few, with every generation materially more capable than the one before it, and improvements are growing daily.
The Capital Being Invested
Running these models requires a physical infrastructure buildout on a scale the world has rarely seen before. The closest comparisons are the railways of the nineteenth century, the electrification of cities in the 1920s and the fibre-optic boom around 2000. Combined capital expenditure across the five largest U.S. operators of data centres, being Amazon, Meta, Alphabet, Microsoft and Oracle, has risen from around US$160 billion in 2022 to a forecast approaching US$1 trillion a year by 2030.
It helps to picture the industry in layers. At the base sit electricity and raw materials, principally copper for wiring and grid expansion, and gas, oil, uranium and renewables for generation. Above that sit the components: processors, memory, cooling systems and electrical equipment. Above those sit the data centres themselves. The research laboratories such as OpenAI and Anthropic rent capacity from those data centres, and at the top sit the businesses we as consumers use and see every day; building products and delivering services using the models.

Source: LSEG Datastream, Resonant Asset Management. Combined capital expenditure, five largest US data centre operators.
Hyperscalers are the U.S. tech giants (Amazon, Meta, Alphabet, Microsoft and Oracle) that own and run the world’s largest data centres. Hyperscaler; a typically eye catching piece of financial jargon, that labels businesses having built an application or service that can experience significant growth in users with only very marginal increase in operational costs. They have no shortage of cash to spend, with the real constraint being the supply chain. The Hyperscalers, however, are competing for scarce electricity, computer memory, advanced chip packaging from TSMC in Taiwan and the most advanced processors from Nvidia. If those shortages eased, the spending would be higher.

Source: LSEG Datastream, Resonant Asset Management. Combined free cash flow, five largest US data centre operators.
We have been monitoring the free cash flow (the cash they have left after paying to run and grow the business) of the big U.S. technology companies, which analysts expect to fall by about 90% between 2024 and 2027 as these companies spend more on AI construction. Can the ‘hyperscalers’ get to a point of sustained, growing profit margins given their upfront capital investments? This is the question that only time will tell. Given the supply chain issues in bedding down their operational frameworks, we have been positioning portfolios away from the U.S. Mega Caps, which are spending a large share of today’s profits for a payoff still many years away.
Where Your Portfolio Is Positioned
Our focus has been on companies in the supply chain that artificial intelligence construction programs depend on, rather than the companies doing the spending. The distinction matters because the suppliers are receiving cash now, while the Hyperscaler companies building the data centres are absorbing the cost of that spending well ahead of any profit return.
- Australian Equities: Australia supplies a meaningful share of physical commodity inputs required for the artificial intelligence build out. Copper is required for data centre wiring and for the electricity network expansion that must accompany it. Both current portfolio overweights, BHP Group and Rio Tinto, have been sensibly directing capital towards growing their copper divisions.
- Global Equities: International portfolios have focused on hardware and semiconductor companies, like SanDisk, Samsung Electronics and SK Hynix, that manufacture the processors and memory the data centre build-out depends on. These businesses are being paid in cash now, and have seen strong upward revisions to earnings, reflecting sustained AI-driven capital investment and profit growth. At the same time, we are underweight the largest names in the benchmark, including many in the Magnificent Seven, which are committing a large share of their cash flow to a build-out where the payoff is still many years away.
- Global Infrastructure: A large data centre uses as much electricity as a small city, running 24/7. Meeting that demand requires new generation, grid and transmission capacity, and the spending flows to the regulated electric utilities and grid operators that own and run this network, who get paid regardless of which AI company ends up winning. Their revenues are set by regulators and backed by long-term contracts, giving steady, predictable cash flows and a lower-risk way to participate in the AI build-out. Our meaningful exposures here are Entergy, NextEra Energy, Pinnacle West and Public Service Enterprise Group.

Conclusion
At Vantage, our investment philosophy is to own quality businesses at sensible valuations, build well-diversified portfolios and invest for the long term rather than chasing short-term thematics. Staying measured in being ahead of the game, rather than chasing yesterday’s winners. Our aim is to construct portfolios that perform well as rational outcomes emerge, rather than an ‘all-in’ perspective on a single view of a technology being correct.
*This article contains purely factual information and/or general advice and does not constitute personal financial product advice. The content of this article does not take into account your personal objectives, financial situation or needs and you must determine whether it is appropriate to your situation. We recommend you obtain financial, legal and taxation advice before making any financial investment decision.
About the Author:

Anthony Nguyen – Head of Investments – Vantage Wealth Management
Anthony commenced his career in the financial services industry in 2018, bringing a strong educational background to his role. He holds a Bachelor of Commerce from the University of Western Australia and a Master of Finance from Curtin University, where he received the Curtin School of Economics and Finance Head of School Prize for graduating at the top of his class. He has also successfully passed the CFA Level 2 exam.
As the Head of Investment at Vantage Wealth Management, Anthony is responsible for overseeing investment due diligence, investment governance and portfolio management of the Vantage Managed Accounts. He is an integral member of both the Vantage Investment Team and the Vantage Investment Committee.
