Three years after OpenAI ignited global enthusiasm for artificial intelligence with the launch of ChatGPT, investor confidence is showing signs of strain. While funding continues to pour into the sector, doubts are growing about whether current spending levels can deliver sustainable returns.
Recent market moves have added to the unease. Nvidia shares have sold off, Oracle stock fell after reporting higher-than-expected AI-related spending, and sentiment has cooled around companies closely linked to OpenAI. As investors look ahead to 2026, debate is intensifying over whether to scale back AI exposure or continue betting on a technology widely viewed as transformative.
“We’re in the phase of the cycle where the rubber meets the road,” said Jim Morrow, chief executive officer of Callodine Capital Management. “It’s been a good story, but now we’re testing whether the returns will justify the investment.”
Much of the concern centers on the cost of building and running AI systems, and whether customers will ultimately pay enough to support those investments. The S&P 500’s $30 trillion bull run over the past three years has been driven largely by Big Tech and AI infrastructure beneficiaries such as Alphabet, Microsoft, Nvidia and Broadcom. If growth in those stocks slows, broader markets could follow.
OpenAI alone plans to spend as much as $1.4 trillion in the coming years, despite generating far less revenue than its operating costs. The company is expected to burn through $115 billion by 2029 before turning cash-flow positive in 2030, according to reports cited in the analysis. While fundraising has remained strong so far, investors are increasingly wary of “circular financing,” where chipmakers and infrastructure providers invest heavily in the same customers buying their products.
Spending pressures extend beyond startups. Alphabet, Microsoft, Amazon and Meta are projected to invest more than $400 billion in capital expenditures over the next 12 months, largely for data centers. Depreciation costs tied to that expansion are rising sharply and could weigh on free cash flow, dividends and share buybacks in coming years.
Despite the concerns, analysts caution against drawing direct parallels with the dot-com crash. Valuations for most major AI players remain far below the extremes seen in the early 2000s, and many leading firms are profitable with deep balance sheets.
“This isn’t dot-com multiples,” said Tony DeSpirito, global chief investment officer at BlackRock, noting that while pockets of speculation exist, the largest AI-linked companies are not uniformly overvalued. pasted
For investors, the challenge is navigating a sector caught between genuine long-term potential and near-term financial strain. While a sharp collapse may be unlikely, many expect a rotation away from the most crowded AI trades as markets reassess growth expectations.
