AI Bubble Alert: Experts Warn of Inflated Valuations and Overexposure to Tech Stocks

Investors are questioning whether the AI boom is a bubble, similar to the dot-com bubble of the late '90s. OpenAI CEO Sam Altman described the AI craze as a bubble, and some signs of sliding among tech stocks. Despite this, AI-focused ETFs continue to be listed, and expert outlooks are mixed. Experts warn of inflated valuations and a universal belief that "this time is different." Indicators of a bubble include the dominance of a few tech stocks in the S&P 500 and a recent MIT report finding that most AI pilot programs have no effect on revenue.

Investors are questioning whether the current AI boom is a bubble, similar to the dot-com bubble of the late '90s. OpenAI CEO Sam Altman described the AI craze as a bubble, and some signs of sliding among tech stocks have emerged. Despite this, AI-focused ETFs continue to be listed, and expert outlooks are mixed. Experts warn of inflated valuations and a universal belief that "this time is different." Indicators of a bubble include the dominance of a few tech stocks in the S&P 500 and a recent MIT report finding that most AI pilot programs have no effect on revenue [1].

The MIT report, which sparked a downturn in AI stocks on Wall Street, revealed that 95% of companies see no financial returns from their AI initiatives, with only 5% achieving measurable results [1]. This stark finding has rattled investors, prompting a sell-off in popular AI-related stocks. The report highlights key reasons for this widespread failure, including the rush to deploy AI before establishing robust data pipelines, security measures, or employee training programs. Significant investments are poured into servers and models, yet integrating these technologies into business processes proves to be a slow and costly endeavor [1].

Amid growing chatter about an “AI bubble,” investment funds are pulling out of high-flying AI stocks, exacerbating the market decline. However, this downturn may not signal a collapse but rather a reality check for the industry. The long-term growth of AI will hinge on tangible economic factors, such as reducing inference costs and enhancing model productivity [1]. While the market reacts to the report’s sobering data, it also opens the door for a more grounded approach to AI adoption. Companies that address these foundational issues could lead the next wave of innovation, provided the technology delivers on its economic promises [1].

Microsoft AI (MAI) has introduced two new in-house AI models as part of its mission to create AI that empowers everyone globally [2]. The company released MAI-Voice-1, a speech generation model that can create a full minute of audio in under a second on a single GPU. This model is already powering features in Copilot Daily and Podcasts, and is now available in Copilot Labs where users can test expressive speech and storytelling capabilities. MAI also began public testing of MAI-1-preview on LMArena, a platform for community model evaluation [2].

Institutional investors are betting on Ethereum to outshine Bitcoin in the near term. The United States spot Ethereum (ETH) exchange-traded funds (ETFs) recorded the highest cash net inflows of about $455 million since August 15, 2025. BlackRock’s ETHA led in net cash inflows on Tuesday of about $323 million, thus currently holding about $16.9 billion in net assets [3]. The demand for spot Ether ETFs has outpaced that of spot Bitcoin ETFs in the recent past. The ETH price surged 2 percent during the past 24 hours to trade at about $4,640 on Wednesday during the mid New York session [3].

NVIDIA Corporation (NASDAQ:NVDA) set a new milestone in its latest quarter, reporting $46.7 billion in revenue, above its outlook, with strong momentum across data center, gaming, and networking segments [4]. The company credited the ramp of its Blackwell platform, with the GB300 now in production and adoption of the GB200 NVL system expanding among cloud providers and AI leaders such as OpenAI, Meta Platforms (META), and Mistral. Data center sales climbed 56% year-over-year, even as H20 revenue slipped by $4 billion due to U.S. licensing reviews tied to China. CEO Jensen Huang emphasized the broader opportunity, projecting global AI infrastructure spending of $3 trillion to $4 trillion by the end of the decade [4].

References:
[1] https://quasa.io/media/wall-street-ai-stocks-slide-triggered-by-mit-report
[2] https://www.investing.com/news/stock-market-news/microsoft-ai-unveils-maivoice1-and-mai1preview-models-93CH-4215309
[3] https://coinpedia.org/price-analysis/spot-eth-etfs-records-455m-inflows-ethereum-price-up-2-today/
[4] https://finance.yahoo.com/news/nvda-nvidia-sees-3-4-123235970.html