HSBC Flags AI Capex Mismatch; Others Warn of ‘Irrational Exuberance’

HONG KONG – As the artificial intelligence (AI) boom continues to dominate headlines and corporate strategies, a crucial question looms: Is the massive investment in AI infrastructure translating into commensurate revenue growth? HSBC CEO Georges Elhedery addressed this concern during the Global Financial Leaders’ Investment Summit in Hong Kong, highlighting a potential disconnect between current expenditure and future returns.

Elhedery emphasized that while the demand for AI’s computational power is undeniable, the sheer scale of investment requires careful consideration. He questioned whether current revenue streams are sufficient to justify the substantial capital outlays required to support these advanced technologies.

Industry analysis supports Elhedery’s concerns. Morgan Stanley projected earlier this year that global data center capacity would increase sixfold over the next five years, with an estimated $3 trillion investment in data centers and associated hardware by the end of 2028. McKinsey & Company took it a step further, releasing a report that estimated a staggering $5.2 trillion in capital expenditure would be needed by 2030 to equip data centers for AI processing loads, dwarfing the $1.5 trillion projected for traditional IT applications.

The crux of the issue, according to Elhedery, lies in consumer readiness to pay for AI-driven services and the timeline for realizing tangible productivity gains. He cautioned that businesses might initially be hesitant, as significant productivity improvements are not expected to materialize immediately. “These are like five year trends, and therefore the ramp up means that we will start seeing real revenue benefits and real readiness to pay for it, probably later than than the expectations of investors,” he stated.

William Ford, chairman and CEO of General Atlantic, echoed Elhedery’s perspective. While acknowledging the transformative potential of AI to create entirely new industries and applications, Ford emphasized the long-term nature of the investment horizon. “In the long term, you’re going to create a whole new set of industries and applications, and there will be a productivity payoff, but that’s a 10-, 20-year play,” Ford said at the summit’s panel.

The capital expenditure forecasts of Big Tech companies underline the scale of investment in AI. Alphabet, Meta, Microsoft, and Amazon have collectively revised their capital expenditure budgets upwards, now anticipating a total of over $380 billion this year. OpenAI, the company behind ChatGPT, has also forged substantial infrastructure agreements worth approximately $1 trillion with industry giants Nvidia, Oracle, and Broadcom.

Ford notes that such immense investment indicates recognition of AI’s long-term potential, but that the sector is initially capital-intensive, requiring significant upfront investment for future returns. He also cautioned against potential pitfalls inherent in the early stages of this technological revolution. “You need to, sort of, pay up front for the opportunity that’s going to come down the road,” he stated but warned that there could be “misallocation of capital, destruction, overvaluation… [and] irrational exuberance” in the initial stages, and also added that it can be difficult to pick winners and losers at the moment.

Ford draws parallels to transformative technologies of the past, such as railroads and electricity. “You’re really betting on this being a broad based technology, more like railroads or electricity, that had profound impacts over over time, and reshaped the economy, but were very hard to predict exactly how in the first few years.” The comments highlight the inherent uncertainties and long-term outlook needed to navigate the AI landscape.