Four leading chatbots offered broadly similar but distinct views on whether the artificial intelligence industry is in a bubble and what could happen if investment enthusiasm fades. ChatGPT, Gemini, Claude and Perplexity all identified risks in areas such as startup funding, infrastructure spending and company valuations, while rejecting the idea that AI itself is necessarily a temporary fad.
ChatGPT separated the usefulness of AI technology from the possibility that investors are paying too much for it. It argued that AI already supports writing, coding, analysis, automation and content creation, but warned that companies can build valuable products while markets overestimate how quickly those products will generate profits. The chatbot compared this possibility with the dot-com era, when the internet transformed the economy even though many internet companies collapsed after their valuations moved ahead of their businesses.
According to ChatGPT, a serious reckoning could occur between 2027 and 2029 if industry revenue fails to keep pace with private infrastructure spending or customers find that AI delivers less measurable value than expected. It also stressed that no one can reliably predict whether or when a bubble will burst.
Gemini highlighted several parallels with the dot-com crash. It compared today’s expansion of GPU clusters and data centers with the overbuilding of fiber-optic networks in the late 1990s, when demand took longer than expected to catch up. It also pointed to circular spending patterns, in which AI startups use investment funds to purchase cloud capacity and computing resources, as well as expectations that may be moving faster than real-world monetization.
Gemini nevertheless noted an important difference between the two periods. Major technology companies supporting AI have other substantial businesses and investments. It argued that companies such as Amazon and Microsoft could face financial pressure if the AI market weakened, but would be less likely to disappear entirely than many dot-com companies were after the 2000–2002 crash.
Claude focused on the scale of current spending and the financial strength of the companies leading the investment. It cited projected 2026 spending of at least $700 billion by Meta, Microsoft, Alphabet and Amazon, describing this as a major source of concern. At the same time, it said large cloud providers have stronger balance sheets than many companies involved in the dot-com boom and that demand for AI capacity remains substantial.
Claude identified several possible triggers for a downturn, including higher interest rates, disappointing returns on AI investment, credit-market problems and cheaper competition from Chinese AI models. One advisory firm reportedly identified 2027 as its best estimate for a possible break, while Capital Economics also expects a correction in AI-related stocks. Claude emphasized that these remain forecasts rather than reliable predictions.
Perplexity described the situation as “pockets of speculative excess around a real technology” rather than evidence that the entire industry is overvalued. It distinguished a stock-market decline from a broader debt crisis, noting that the latter would have wider economic consequences. The chatbot also said that a downturn could result from several problems arriving together, such as weak earnings, spending reductions and credit difficulties.
Across the four responses, the main point of agreement was that some parts of the AI market appear vulnerable. However, none of the chatbots could establish that a sector-wide crash is imminent or provide a dependable date. Their assessments instead centered on whether AI-generated revenue and measurable business value will eventually justify the money being spent on infrastructure, companies and related investments.
Sources: Tom's Guide.