The rise of artificial intelligence over the last three years has changed the way businesses, financial markets, and even our daily lives operate. As demand for generative AI models grows, so does the need for data centres, computational power, and most importantly, electricity. The challenge that faces us is innovating faster than ever before to keep up with the trend.
Rapid Industry Growth
OpenAI’s first ChatGPT model launched in November 2022. Within two months, it became one of the fastest-growing apps in history, reaching 100 million users and marking a turning point for the industry. Since then, there has been widespread adoption and investment in new models, companies, and infrastructure.
Statista estimates the entire AI market at $254 billion, up from $142 billion in 2022 when ChatGPT launched. This represents roughly 21% CAGR. Forecasts suggest the industry could exceed $1.8 trillion by 2030, implying a sustained 37% CAGR over the eight years.
If these numbers hold, they point to the monumental expansion of the sector but also raise questions about the energy consumption that comes with it. Machine learning and model training require significant computational power, and this power comes from energy-heavy data centres filled with high-speed processors.
The Infrastructure Challenge
According to the International Energy Agency (IEA), global data centres are expected to demand almost 1,000 TWh of electricity by 2030, up sharply from around 425 TWh today.
This rising demand is expected to test the limits of global electricity grids, requiring major investments in new generation capacity, grid management, and energy storage. To sustain this trajectory, technology must evolve rapidly, with advances in chip efficiency, cooling systems, and model optimisation.
Here lies the opportunity. Companies driving technological improvements will offer the most value to consumers and investors. Any business delivering meaningful efficiency gains or electricity savings will lead the pack and, hopefully, accelerate further innovation.
For the public, the impact may be harsh. Data centres must be built where grid infrastructure already exists, often near residential areas, creating a simple supply-and-demand problem. Areas with data centres will require more power, meaning more capacity must be added to the grid, putting upward pressure on household electricity bills.
Our environment will also feel the effect if left unchecked. The IMF estimates AI-related energy demand could add 1.7 gigatons of CO₂ over the next five years if not managed, offsetting many efforts toward carbon neutrality.
Who Are the Winners?
It’s another sink-or-swim situation. Rising energy demand presents both risks and opportunities across multiple sectors. Companies operating in energy production, renewables, storage, transmission, and component manufacturing are set to benefit. Semiconductor firms producing advanced GPUs and other processors have already seen significant gains in both earnings and share prices.
Two household names come to mind for investors: Nvidia (+1,310%) and Taiwan Semiconductor (+370%) since ChatGPT’s launch. It’s clear that the adoption of AI has resulted in monumental gains for these companies. For South African investors, the opportunities may still be around the corner.
South Africa is a resource-based economy with large deposits of minerals, ores, and precious metals still to be mined. Current production relies heavily on metals like copper, gold, silver, and PGMs. Local companies that mine or refine these metals stand to benefit from the increased demand for AI infrastructure.
In addition to our resource market, the vast open land across South Africa could serve as prime locations for renewable energy plants. Whether solar, wind, hydro, or nuclear, this opportunity could position South Africa as both a data and power hub for Africa, adding a new layer of diversification to our economy.
Innovation or Stagnation
Sustainability is the real issue. The increased demand will require the world to innovate and adapt faster than ever before. Generation and storage capacity must increase, grids must expand while becoming more flexible, and AI hardware and software must become far more efficient.
This technology has already changed almost every aspect of our lives and will continue to do so. Optasia, the JSE’s latest public offering, is an AI-fintech company offering microfinancing and airtime credit solutions across 38 countries, with its listing date set for the 3rd of November. The company serves more than 100 million users and 13 financial institutions. To highlight its significance, FNB, issued a report describing the company as “divinely visionary.”
If you want to secure an allocation prior to the listing, contact Finova before 28th October when the offer period is up.
The message is clear for the public, policymakers, and business leaders. The AI boom goes hand in hand with rapid technological development in energy. Companies and countries that succeed will need to produce cleaner, more efficient, and affordable energy at a rate never seen before, while those that fail will bear the costs. The future of artificial intelligence, and our livelihoods, depends heavily on how well we manage the next few years.
