When "retirement funds" meet the "AI boom": What are the risks of global pension funds betting heavily on data center infrastructure?

```

As the artificial intelligence (AI) boom sweeps global financial markets, pension funds and insurance capital that have long sought stable returns are rapidly pouring into data centers and other AI infrastructure sectors, attempting to seize a share in this emerging wave of technology. However, this has sparked deep concerns in the market over the safety of public savings.

Recently, SoftBank Group executives revealed that its $500 billion “Stargate” data center project in the United States may potentially be backed by long-term investors, including life insurance companies and pension funds. SoftBank founder Masayoshi Son holds an extremely optimistic view, believing that considering AI’s immense future contribution to the global economy, even massive investments can be quickly recouped.

Meanwhile, the Canada Pension Plan Investment Board has also announced a cooperation plan with Australia’s Goodman Group, with both parties joining forces to invest billions of dollars in European data center projects. Analysts believe these moves show that, despite the uncertain future profitability of AI technology, tech giants in urgent need of funding for costly computing facilities are perfectly matched with institutional capital seeking long-term asset allocation.

However, deeply tying pension funds—meant for emergencies—to AI infrastructure still in its investment phase brings potential risks that cannot be ignored. While proponents view this as a necessary infrastructure-level investment, analysts warn that without a clear path to profitability, exposing the public’s retirement savings to the risks of a tech bubble could result in painful lessons reminiscent of past asset bubble bursts.

A certain “alignment” between long-term capital and grand AI ambition

From the perspective of investment term matching, long-term capital and data centers appear to be a “match made in heaven.” The distant future depicted by AI optimists happens to align with the long-duration nature of annuity businesses. The core logic is:while short-term misjudgments may occur, decades-long investment periods are enough to smooth out losses and ultimately achieve reliable returns.

Earlier this month, Masayoshi Son articulated his grand business vision. He calculated that if physical artificial intelligence can create 10% of global Gross Domestic Product (GDP) in ten years—which he believes will then reach about $200 trillion—“even if you invest $10 trillion over the next ten years, you could recoup it in six months.” Based on this logic, he asked:

“Where is the bubble?”

Undoubtedly, machine learning and AI will profoundly impact the global economy for decades to come. In pursuit of this goal, tech companies are currently engaged in large-scale fundraising to build the expensive data centers and power plants required. AI advocates say these projects provide essential infrastructure such as water, roads, bridges, and telecommunications networks. Thus, using project financing to support AI infrastructure is seen as building the physical assets that will underpin future investments.

The echo of history: From Japan’s bubble to private equity concerns

Despite grand ambitions, as Bloomberg columnist Mayumi Negishi pointed out, when it comes to ordinary people’s savings, the most conservative investment strategy should be adopted. She recalled the aftermath of the collapse of Japan’s asset price bubble in the late 1990s: In less than four years, seven life insurance companies went bankrupt in succession, severely impacting the “nest egg” funds of millions of policyholders.

This is precisely why, when SoftBank plans to bring in pension funds and when Canada Pension Plan Investment Board teams up with Goodman Group for large-scale data center investments, market observers are wary. Over the past decade, Wall Street capital’s takeover of life insurance companies has produced many cautionary tales.

A typical case is Jimmy Nappo, who diligently paid premiums for 17 years to ensure his wife and daughter’s future. But after his death, the private equity-owned insurance company he was insured with went bankrupt due to capital shortages, and his widow is still fighting for full compensation. These historical experiences show that when long-term savings are introduced into high-risk or high-leverage investment tools, it is often ordinary savers who end up bearing the risk.

Huge investments and the unproven business cycle

Currently, the main reason for large-scale data center construction is efficiency—the larger the scale, the greater the processing power and the lower the unit computing cost. However, the market has yet to see a well-argued roadmap clearly demonstrating how to transition from the current investment frenzy to profits sufficient to cover these enormous costs.

Masayoshi Son’s vision is tempting but is criticized as not being based on real-world mathematics. The industry is currently built on a huge assumption: that the computing power of AI data centers will be as valuable as electricity—enough to cover capital costs and generate surplus. While life insurance companies and pension funds may use debt or preferred shares to control risk and ensure priority payouts in adverse situations, the scale of investment remains alarming.

OpenAI’s Sam Altman has mentioned spending plans of several trillion dollars and related financial instruments; Nvidia’s Jensen Huang has also assertively emphasized the certainty of the AI era. These remarks are influencing investment managers in the financial sector. This inevitably evokes memories of the infrastructure over-investment boom just before the dot-com bubble burst. Although in the long term, technology bubbles and the infrastructure they leave behind may benefit society and improve human lives, it does not mean investing personal pension funds in them is wise. For investors who rely primarily on pensions for living, avoiding such uncertain grand wagers may be the safer choice.

Risk Warning and DisclaimerThe market has risks; investment must be cautious. This article does not constitute personal investment advice and does not take into account individual users’ specific investment objectives, financial situation, or needs. Users should consider whether any opinions, views, or conclusions in this article are suitable for their particular circumstances. Investment based on this is at your own risk. ```