The Emerging Techno And Energy Economic Race
Gearing up for the wider adoption of artificial intelligence (AI), companies operating data centres are poised to invest over $100 billion in expanding their capacity in Asia over the next few years. This significant investment underscores the region's growing importance in the global tech landscape.
However, Australia, driven by its strategic alliances with the US and other Western nations, is also vying to position itself as a key player in this techno-economic race, representing allied interests in the AI technology domain.
This infrastructure expansion is power-intensive. A single ChatGPT request can require ten times more power than a Google search, according to the International Energy Agency. As more computing power comes online, Asian countries are likely to experience a significant surge in electricity demand, potentially doubling in some regions.
The need for robust investment in power grids, which are increasingly becoming bottlenecks, is critical to support this growth. Countries like the U.S. and Singapore have already restricted data centre expansion due to tight power markets, although Singapore plans to resume adding capacity after addressing power generation issues.
The availability of stable baseload power is essential to fuel Asia's computing growth. Proprietary Morgan Stanley Research forecasts a 20% annual growth in data centre capacity in Asia through 2027, with power requirements doubling by 2027 and tripling by 2030 to around 40 gigawatts.
This raises significant questions about the energy needed to fully realise the potential of generative AI, a revolutionary but incredibly power-intensive technology. As AI embeds into more products and services, energy demand is expected to intensify, potentially skyrocketing by 70% annually.
“We were surprised by our own projections showing such a rapid surge in power demand within just a few years,” said Simon Flannery, a Morgan Stanley Research analyst covering North American telecom services and communications infrastructure.
“We think the huge step up in energy needs is not well understood by the market and hasn’t been priced into a number of stocks.”
Japan, Malaysia, and Thailand are attracting significant investments from computing infrastructure companies, known as hyperscalers, due to their reliable power infrastructure. Microsoft and Amazon have recently announced plans to scale up data centres in Southeast Asia, while Nvidia has partnered with Malaysia's YTL Power to ensure stable and clean power for its new AI chip-based data centre.
Southeast Asia is emerging as an attractive destination for data centres and AI-led infrastructure because of its affordable power and land, easy grid connections, and a $50 billion pipeline of planned power-sector investments over the next five years.
Australia's Strategic Positioning
The geopolitical and techno-economic race is not confined to Asia alone. Australia is keenly aware of the strategic implications of AI and advanced computing technologies. With its close ties to the US and a commitment to representing Western interests in the global south, Australia is striving to enhance its tech infrastructure.
The country is only beginning to address the critical issues of energy transition, with nuclear energy emerging as a viable alternative to support the Fourth Industrial Revolution.
The energy debate in Australia is intensifying, focusing on how to sustain industrial growth while transitioning to cleaner energy sources. Nuclear power, despite its controversies, is being reconsidered as a stable and potent energy source to back the nation's AI and technological advancements.
Australia's vast land and resources provide a unique opportunity to develop a diversified energy portfolio that includes renewables and potentially nuclear energy to ensure a stable power supply for its growing tech sector.
Australia faces a critical challenge: will it have the policies in place to commit fully to an energy transition that can manage the high energy requirements needed to support digitization and the democratisation of AI across its economic systems and society? In contrast with Asia's enormous investment commitments in energy alternatives, is Australia prepared to look beyond its decarbonization strategy?
Moreover, how will it address the race it is compelled to engage in due to strategic alliances with the United States and the UK as the global south ally of choice? What will Australia do to support future data centres of 100 megawatts, moving toward gigawatt-scale environments? Are Australia's political leaders strategically assessing this with a ten- to fifteen-year view?
In contrast, Asia's competitive advantage in clean power production is noteworthy. As countries expand their green infrastructure, clean power prices are likely to deflate despite higher financing costs. Clean power generation, complemented by gas-based power, is expected to provide round-the-clock electricity at competitive prices.
Over the past decade, hyperscalers have significantly improved power efficiency, keeping annual electricity consumption growth to single-digit rates. This trend is expected to continue with more liquid-cooled data centres coming online.
The techno-economic race between Australia and Asia in the AI domain is shaping the future of global technology hubs. As Australia competes with Asia to become a key player in the global south, it must address its energy policies and leverage its alliances to drive AI adoption. The region that successfully integrates affordable, reliable power with advanced tech infrastructure will lead the next wave of the digital revolution. Can Australia rise to this challenge?