Discussions about the artificial intelligence bubble did not subside during the final months of 2025, in light of the continued rise in technology stock valuations, amid a wave of excessive enthusiasm about investments and fears of missing out on opportunities in the sector. But at the end of the year, most global banks took a calmer tone on the possibility of a bubble next year, while some warnings came from economists and prominent figures. In the following report, we review the most prominent things that have been said about the valuations of stocks of artificial intelligence companies in 2026. Continued concerns about investment returns. Barclays Private Bank has seen that the controversy surrounding the artificial intelligence bubble revolves around whether current valuations and investment spending are outpacing the rate of returns. Read more: AI bubble fears grip Wall Street But the bank noted that today’s AI leaders differ from the tech companies of the late 1990s in that they are profitable, cash-flow-generating companies closely tied to business infrastructure and consumers. Although valuation risks persist, weaknesses relate more to shrinking multiples or slowing technology adoption, rather than the collapse of business models. Diversification and selectivity remain among the best tools for achieving profits while reducing risks. Amid fears that bets on artificial intelligence will not bear fruit, sector companies must achieve revenues estimated at two trillion dollars by 2030 to meet demand… as shown in the following video: For BlackRock, bubbles accompany all major economic transformations, and the matter may happen again. But she pointed out that the bubbles often grow gradually and don’t become obvious until they burst. The P/E of the seven largest companies does not herald a bubble. For this reason, BlackRock closely monitors the size of investments and potential returns, and considers them the basis of its analysis of the path of transformation led by artificial intelligence. Despite the concerns, the company still takes a supportive stance on risk, viewing artificial intelligence as the primary driver of US company stocks. From the perspective of P/Es, John Belton, portfolio manager at Ghibli Funds, which manages assets worth $35 billion, believes that the artificial intelligence market has not yet entered a valuation bubble, noting that the average P/E for the seven major tech companies (Apple, Microsoft, Amazon, Alphabet, Meta Platforms, Tesla, or N3259 times times when times are excluded, or N259). Tesla, compared to P/Es of 39.9 times. 90 times for companies in late 1999 during the dot-com bubble. You may be interested in: Are we witnessing an artificial intelligence bubble? The bond market does not think so, confirming that these valuations reflect strong fundamentals and are not the result of excessive speculation. Among other reasons Beltone relies on is that high capital spending in the artificial intelligence sector is often directed at strengthening existing and profitable companies, rather than on untested bets, and the use cases are expanding to include areas such as self-driving, robotics, biosciences and generative software, improving the possibilities of achieving real commercial returns, he says. NVIDIA rejects the idea of ​​a bubble. However, he believes that the infrastructure investment cycle will inevitably peak, but the challenge lies in knowing the timing and shape of that peak. Read more: Nvidia’s strong forecasts calm fears of an artificial intelligence bubble. Nvidia also rejected the idea that its activity is part of a bubble, although it acknowledged the presence of concerns in the market. CEO Jensen Huang said that what the company is building is based on real demand and tangible technological transformation, not speculation, adding: “We see a completely different reality than those who describe the market as a bubble.” Are investors buying a dream? Bank of America expects the artificial intelligence sector to experience a “bump” in its road, as the sector could face a slowdown in growth or a rise in prices due to the lack of clarity on the actual path to profitability. Also read: Nvidia’s stock falls amid growing doubts about its dominance in artificial intelligence. Savita Subramanian, head of US equity strategy and quantitative analysis at the bank, said it was still unclear whether gains from artificial intelligence would be achieved. She sees the ability to conserve energy as a major hurdle that takes time to build. “At the moment, investors are buying a dream,” she added. “Goldman” warns against the rise in the sector’s debt, while “Goldman Sachs” focused on the credit aspect, as it explained that since 2022 the technology giants known as “hyperscalers” have financed their spending on artificial intelligence from cash flow, but this trend has begun to change with the rise in expenses and the stability of dividends. This led some of these companies to go to debt markets. See also: Profit-taking pressures Wall Street, hits US artificial intelligence stocks. Through the end of October 2025, five of these companies had issued bonds worth about $90 billion, within the total investment bond market of $1.5 trillion for the same period. The bank expects debt reliance to accelerate as artificial intelligence moves from infrastructure to applications, especially if profits slow and stock dividends don’t fall. Although the financial position of companies remains strong, the rise in debt levels requires monitoring during 2026, according to the bank’s report, which also pointed to the importance of monitoring companies’ ability to achieve tangible returns from their investments. Artificial intelligence valuations will continue to increase in 2026. Capital Economics expects valuations to continue to inflate through 2026 before any major correction occurs later, and warned that the current environment bears many of the hallmarks of a bubble, such as high valuations and exaggerated expectations, while this was most evident in the Bank. technology and artificial intelligence sector is the biggest expected threat to the markets in 2026 based on the results of a recent survey reported by the website. Investing.com. Also read: What are Deutsche Bank’s expectations for global markets in 2026? The German Bank explained that the decline in valuations due to the decline in enthusiasm for artificial intelligence constitutes the first risk next year, noting that US technology stocks are classified among the highest levels of bubble risk. Economist Ruchir Sharma also believes that the rise of artificial intelligence bears all the hallmarks of historical financial bubbles, from overinvestment and exaggeration in valuations to extreme concentration of ownership and resort to borrowing. He pointed out that any increase in interest rates during 2026 could lead to a sharp correction. Bill Gates: Artificial intelligence has the characteristics of a bubble. Bill Gates also warned that artificial intelligence could have the characteristics of a bubble, explaining that not all high valuations in the sector will survive. He said that some companies will see a decline in their value, despite the huge momentum around the technology. Speaking to CNBC on the ninth of this month during Abu Dhabi Finance Week, the founder of Microsoft indicated that artificial intelligence is “the most important thing that is happening right now,” but he added: “It does not mean that all companies with high ratings will win, but rather the field will be very competitive.” He added that AI “is only a bubble in the sense that some of these valuations will not go up, they will go down.” “DeepMind”: A fix is ​​expected at a later stage for artificial intelligence. Demis Hassabis, co-founder and CEO of “DeepMind”, believes that the current state of mania in the financing of artificial intelligence has led to the emergence of bubbles, centered on excessive valuations that can reach tens of billions for some startups, which are still in their early stages and have not yet started their real activity. You may also be interested in: Bezos: The boom in artificial intelligence spending is a “bubble” that will pay off. On the other hand, Hassabis distinguishes between these evaluations and the investments of large technology companies, which he sees as supported by real business. However, an “overcorrection” is expected at a later stage, as usually happens in major technological transformations when the outlook turns from doubt to obsession, according to the “Google DeepMind: The Podcast” podcast, which Business Insider reported on.