While it often feels like Artificial Intelligence (AI) mania has been fuelling markets, it’s worth thinking about what’s powering AI. At the ground level, data centres provide the infrastructure for the raw computing power and storage to support operations.
Right now, the average AI-focused data centre consumes around the same electricity as 100,000 homes. Larger facilities under construction are set to consume 20 times this. Technology giants are grappling with surging emissions as they race to stay competitive and meet the rising demands of digital innovation.
Sourcing a reliable, clean and affordable supply of power is a significant challenge for the industry. But with demand ballooning, capturing sustainable growth will also hinge on the energy efficiency of the AI chain.
This article isn’t personal advice. If you’re not sure an investment is right for you, seek advice. Investments and any income from them will rise and fall in value, so you could get back less than you invest. Ratios also shouldn’t be looked at on their own.
Investing in an individual company isn’t right for everyone because if that company fails, you could lose your whole investment. If you cannot afford this, investing in a single company might not be right for you. You should make sure you understand the companies you’re investing in and their specific risks. You should also make sure any shares you own are part of a diversified portfolio.
The race for efficiency
In 2024, the green economy generated over $5trn in annual revenue. With almost half made up of Energy Management and Efficiency activities, like green buildings, cloud computing and efficient power electronics.
High efficiency semiconductor chips are one way for investors to gain exposure to these solutions.
As well as conserving energy, more efficient semiconductors potentially unlock a better return on investment for AI systems. The current ‘blitzscaling’ approach of extreme levels of investment, chip stockpiling, and high-value talent acquisition has propelled the industry to new heights. But a resource hungry race to the top, focussed on electricity load, investment and accumulation might prove unsustainable for more than just the environment.
While semiconductors have benefited from this boom, a sustainable, resource-based approach to AI might prove to be the industry’s holy grail. For example, a sharper focus on intellectual output for every unit of energy used could pave a more strategic and value-driven path forward for the industry.
What’s next for the sector?
Technology giants like Nvidia and Taiwan Semiconductor Manufacturing Company (TSMC) are leading the charge to make AI computing not only faster but also much more energy efficient.
Nvidia – the designer
Nvidia designs the chips that run most of today’s AI systems. Its latest Blackwell chips, are built for speed and efficiency. Compared to its predecessor, Blackwell can deliver more computing power while using far less energy per task.
There’s a lot going on under the hood. Blackwell introduces a new way of handling AI calculations that reduces the amount of energy needed without sacrificing accuracy. It also improves how chips talk to each other. AI models are huge, often using multiple chips working together. Blackwell’s high-speed connections mean less energy wasted moving data around.
That’s just the tip of the technological iceberg, but the result is up to 25 times better energy efficiency for some AI tasks compared to older systems. A big deal for data centres.
TSMC – the builder
While Nvidia designs the chips, TSMC makes them. TSMC is the world’s leading chip manufacturer, and its job is to make chips smaller, faster, and more efficient.
TSMC brings designs to life using advanced manufacturing processes that pack billions of transistors (tiny electrical components) onto silicon wafers, layer by layer. The process is extremely complex, using extreme ultraviolet (EUV) light and nanometre-scale precision.
Smaller transistors use less energy, and TSMC’s latest processes improve power efficiency by 22% over earlier technology, while its upcoming projects will see similarly impressive improvements.
Why all this matters
AI is here to stay, but its energy appetite doesn’t have to spiral out of control. Thanks to breakthroughs from companies like this, the next generation of AI hardware will be smarter, faster, and greener. That means we can keep pushing the boundaries of what AI can do, without blowing up the world’s electricity bill.
When we think about AI hardware and chip design and manufacturing, we think energy efficiency will be as important as raw performance in the years to come. Those able to bridge the gap between the two should be best placed to take and retain market share.
The author holds shares in NVIDIA.
This article is original Hargreaves Lansdown content, published by Hargreaves Lansdown. It was correct as at the date of publication, and our views may have changed since then. Unless otherwise stated estimates, including prospective yields, are a consensus of analyst forecasts provided by LSEG. These estimates are not a reliable indicator of future performance. Past performance is not a guide to the future. Investments rise and fall in value so investors could make a loss. Yields are variable and not guaranteed.
This article is not advice or a recommendation to buy, sell or hold any investment. No view is given on the present or future value or price of any investment, and investors should form their own view on any proposed investment. This article has not been prepared in accordance with legal requirements designed to promote the independence of investment research and is considered a marketing communication. Non-independent research is not subject to FCA rules prohibiting dealing ahead of research, however HL has put controls in place (including dealing restrictions, physical and information barriers) to manage potential conflicts of interest presented by such dealing. Please see our full non-independent research disclosure for more information.
