Investment Perspective | How AI is creating new opportunities in commodities

Artificial intelligence (AI) has been one of the most powerful investment themes in recent years. While much of the focus remains concentrated on semiconductors and the software layer, behind the rapid buildout of AI lies a far more physical reality.

The proliferation of AI data centres is driving strong demand for commodities, from copper and industrial metals to energy and power-related inputs. At the same time, supply-side constraints are becoming more apparent.

As Paul Gooden, Head of Natural Resources and Co‑Portfolio Manager for the Global Natural Resources strategy at Ninety One, observes, there is a there is a growing realisation that “the ethereal world of the tech bros needs the material world of what the natural resource companies do.

In this discussion below hosted by Shankar Panchadcharam, Senior Director, Business Development at UOBAM, find out how these trends are unfolding and what they could mean for investors looking to add commodities to their portfolios.

Transcript

Are we in a commodity supercycle?

Shankar Panchadcharam: Hi Paul, there's been increasing discussions on a potential new commodity supercycle. It's driven by a few things, like structural demand and supply constraints. What do you think about the debate?

Paul Gooden: The commodity supercycle that you're referencing in the early 2000s was really driven by industrialisation and urbanisation in China, leading to a broad-based demand-pull commodity cycle.

I think we are in the early stages of a commodity cycle, but I think it's a little bit more nuanced and a little bit different to the early 2000s.

There are two sorts of differences. The first one is that we're moving into a multipolar world, where traditional supply chains are breaking down due to tariffs, sanctions, and wars. And as a result, there's more resource nationalism and countries will likely want to build domestic stockpiles and inventories. So, the apparent demand is going to be higher than the underlying demand because you're going to have this kind of stock building.

The second thing is capital discipline in these industries is pretty strong, so the supply side is quite constrained. You put it all together, and yes, I think you can see a commodity cycle developing. I think we're in the relatively early stages of it.

But I wouldn't call it a commodity supercycle. I think, as I said, it's more nuanced. You could possibly call it the revenge of the old economy, or you could call it a realisation that the ethereal world of the tech bros needs the material world of what the natural resource companies do.

Outlook and opportunities

Shankar Panchadcharam: Energy has done relatively well, and that's driven by electrical demand and ongoing political uncertainty. What's your outlook for this sector and especially for the second half and beyond?

Paul Gooden: In terms of energy, the most obvious issue at the moment is what is going on in the Strait of Hormuz. The outage is about 12 million barrels a day¹. So it's very meaningful.

I would say you can divide it into two timeframes. There are the coming days, weeks, months, we're waiting on a deal, and a deal will happen because it has to happen. Because the global economy will go into recession if there isn't a deal.

And then I think for me as an equity investor, what's more interesting is the medium term. I think the mid-cycle oil price is going to move higher. Before this crisis at the start of the year, I thought US$70 Brent was a reasonable mid-cycle price. I now think US$80 is more reasonable².

At the end of this crisis, we would have drawn down maybe a billion and a half barrels of inventories. And if you want to rebuild those lost stockpiles over 3 years, it’s more than a million barrels a day of incremental oil demand. And so, it means the underlying oil demand is going to be more than double what it's been in recent years.

And that incremental demand is coming at a time when there is very strong capital discipline in the industry. In particular, US shale is running into tougher geology. So I think you need meaningfully higher oil prices to incentivise the supply to satisfy demand.

Shankar Panchadcharam: Beyond energy, do you see any notable opportunities, especially within the metals and agriculture space?

Paul Gooden: Two that stand out are fertilisers and aluminium, both very energy intensive. About a third of the world's nitrogen-based fertiliser normally flows through the Strait of Hormuz and it's kind of blocked in. About 78 percent of the world's aluminium supply comes from the Middle East, and about half of that aluminium supply is blocked in from Hormuz. So we think both fertilisers and aluminium look quite interesting at the moment.

The case for natural resource equities

Shankar Panchadcharam: Paul, natural resources is a really interesting asset class. Why should investors consider buying natural resource equities as opposed to physical assets?

Paul Gooden: I would say on the commodity side, the fact that the underlying demand outlook is okay and you've got strong capital discipline means that the commodity market should tighten over time. Natural resource equities, to a large extent, are plays on those commodity prices, but equities I think, should broadly benefit from other trends going on in equities.

So, one is structurally higher inflation. In periods of high inflation, few asset classes perform as well as natural resource equities.

Another dynamic in public markets is the HALO trade (Hard Assets, Lower Obsolescence).

So, to the extent that investors are worried about AI disrupting business models, it's difficult to disrupt these business models. Because of the incremental demand for metals and power, AI actually is a net positive for this space.

As the cycle kind of rolls on, typically what happens is you get a relaxation in the capital discipline. Companies get seduced by the high commodity prices and they do start to FID (final investment decision) projects.

And I would say for the most part, we're not seeing that. We're seeing good capital discipline. And so that's another reason that could give me confidence that we're in the relatively early stages here.

Shankar Panchadcharam: When it comes to investments, there’s always this concern about valuations. You see sky high valuations in AI, tech sectors. Where are valuations in natural resources?

Paul Gooden: When I look at our portfolio in the natural resources fund, the average kind of free cash flow yield of our holdings is around 6 – 7 percent³. So materially above the broader market. And importantly, balance sheets of these companies are pretty strong, capital discipline is very strong as well. Most of that free cash flow is coming back to shareholders.

The distribution yield of a typical stock in our portfolio is about 5 percent through a combination of buybacks and dividends⁴. So valuation still looks pretty attractive.

Why an active approach can add value

Shankar Panchadcharam: So, how should an investor approach natural resources? Do investors go and buy individual securities? Do they buy ETFs? Is there more value towards an all-encompassing portfolio where you're a little bit more dynamic because ultimately not everything runs at the same time and there's a need to basically be a little bit more dynamic in the allocation.

Paul Gooden⁵: If you look over a long-term period, the data shows that you're better off owning equities than the underlying physical commodities. And the reason for that, as you reference, is that different commodities move in slightly different cycles.

So if I look at our natural resources fund, last year, very strong performance driven mostly by precious metals and copper. And the fund has shifted a bit. We're now more overweight energy, overweight fertilisers and aluminium. So, yes, that dynamic approach is very helpful.

And when I go back and look at the attribution of the performance in the fund since inception⁶, it's two-thirds from idiosyncratic company specific stock selection, and about one-third from allocation to different sub-sectors. So absolutely yes, I think an active approach here is very useful.

Gold’s long-term appeal

Shankar Panchadcharam: Paul, one of the commodities that people talk a lot about is obviously gold. Where do you see gold equities, especially in this cycle?

Paul Gooden⁷: In terms of gold equities, as I mentioned, we were very overweight last year. We're now more equal weight gold equities. And to answer that question, we need to look at gold.

Long term, we are structurally bullish gold. Firstly, when you look at the thematic drivers, they are mostly intact. Gold's a good place to be in times of geopolitical risk.

Secondly, the US dollar, our view is we're entering a dollar downside. And that is good for gold.

Thirdly, there are plenty of signs that governments are still running big fiscal deficits. They're running the economy hot, and gold is a good inflation hedge.

And then, when you look at the positioning in gold, central banks have been buying in the last few years, but if you look at the overall central bank reserves in gold, it's about 30 percent. And back in 1980, it was like 70 percent. So there's still plenty of room for central banks to add gold.

So I would say on gold, we're structurally bullish, but at the moment within the global natural resources fund, we are more neutral weight and we are playing other commodities more overweight.

Key investor takeaways

Shankar Panchadcharam: Thanks Paul. So let me take a step back. We’ve talked about a couple of structural trends. What are the three or four key takeaways for investors if they were to invest in the asset class?

Paul Gooden: I would say in terms of the structural investment case, it is kind of multifaceted. They're all long duration themes.

The first one is around electrification. It’s a big global megatrend driving demand for copper, aluminium, and power. That trend has got plenty further to run, particularly as the AI data centre theme gathers pace.

And secondly, it's around inflation. Our view as a house is that inflation is going to be above trend for the medium term. And in periods of high inflation, few asset classes perform as well as natural resource equities. Central banks can print currency, but you can't print molecules or electrons.

Thirdly, this is quite an uncorrelated asset class compared to many other equity strategies out there. As an investor that wants to diversify, this is a very useful place to be.

Fourthly, the HALO trade. If you're concerned about AI impacting some parts of your portfolio, this is a good place to be.

And performance, I guess, is the final thing. Not just the performance of our fund versus our benchmark, which is very strong, but also the performance of our equities benchmark versus the physical commodities is strong as well8.

¹Source: Ninety One, as of 28 May 2026
²Source: Ninety One, as of 28 May 2026
³Source: Ninety One, as of 28 May 2026 . Past performance is not indicative of future performance.
⁴Source: Ninety One, as of 28 May 2026 . Past performance is not indicative of future performance/
⁵Portfolio manager views and strategy as of 28 May 2026
⁶Fund inception: 29 May 2006
⁷Portfolio manager views and strategy as of 28 May 2026
⁸Source: Ninety One, as of 28 May 2026. Past performance is not indicative of future performance