Google chief warns of 'irrational' surges as AI investment boom stirs bubble fears

'I think no company is going to be immune, including us,' Pichai says

Sundar Pichai, chief executive of Google's parent company Alphabet, has warned that the trillion-dollar investment boom enveloping the AI sector carries "elements of irrationality," and that no company would be insulated if the bubble were to burst.

Speaking to the BBC, Pichai described the current wave of AI spending as an "extraordinary moment," but one that echoes previous overheated periods in tech investing.

"I think no company is going to be immune, including us," he said.

The comments arrive amid growing concern in global markets that AI enthusiasm may be reaching an unsustainable peak.

Alphabet's shares have doubled over the past seven months, lifting its valuation to $3.5 trillion.

Nvidia recently became the world's first $5 trillion company. And yet, analysts are increasingly uneasy about the sheer scale of capital flowing into AI ventures.

A particularly scrutinised figure: roughly $1.4 trillion in deals orbiting OpenAI, despite forecasts that the company's revenues this year will amount to less than one-thousandth of that sum.

Comparisons to the late-1990s dot-com bubble have grown louder, with some investors questioning whether the industry is laying far more financial track than near-term economic returns can justify.

Pichai appeared to acknowledge those parallels. Recalling former Federal Reserve chair Alan Greenspan's famous warning about "irrational exuberance," he noted that the internet era had also seen rampant overinvestment.

"None of us would question whether the internet was profound," he said. "I expect AI to be the same."

Fund managers warn over spending boom

A Bank of America survey this month found that a net 20% of fund managers believe companies are overspending.

More than half said AI equities are already in a bubble, and 45% cited the AI sector as the biggest "tail risk" for global markets, surpassing inflation and weakening consumer demand.

The tech-heavy Nasdaq Composite has retreated more than 5% this month, shaken in part by anxieties tied to AI-related capital expenditure.

Nvidia shares slipped after filings showed Peter Thiel's hedge fund had dumped its entire position ahead of the company's upcoming third-quarter earnings.

Credit markets are also absorbing the shockwave: US firms have issued more than $200 billion in bonds this year to support AI buildouts.

Analysts warn of a funding "flood" ahead as hyperscalers continue aggressive spending.

Barclays estimates AI investments by big tech and smaller players could exceed 10% of US GDP by 2029.

Google's 'full-stack' bet

Pichai noted that Google's advantage lies in controlling its "full stack" of technology: its own AI superchips, massive datasets from platforms like YouTube, and the frontier research powering its generative models.

That integrated approach, he argued, positions Alphabet to weather volatility better than rivals. Even so, he stressed that no company is isolated from systemic shifts.

AI's global electricity consumption is estimated at 1.5% of total usage, and Pichai said the figure will rise sharply without massive investment in new power generation and grid infrastructure.

Alphabet has acknowledged slippage on some climate goals due to the energy intensity of model training, though Pichai reaffirmed the company's 2030 net-zero target.

Calling AI "the most profound technology" humanity has worked on, Pichai said the shift will reshape the labour market more deeply than previous technological revolutions.

"We will have to work through societal disruptions," he said, while insisting the technology will ultimately "create new opportunities."

Computing says

It’s hard not to see Pichai’s comments as an unsubtle pitch for US government support should the AI bubble burst. In the same way that some banks became too big to fail during the 2008 financial crash, arguing that if they went under they’d take the rest of the economy with them, we expect to see similar claims from Big Tech should the current round of circular deals and data centre and power infrastructure buildouts fail to deliver AI profits in time.