What is special about this MIT report that triggered the sharp decline in the US stock market?

A report from the Massachusetts Institute of Technology (MIT) reveals the harsh reality of corporate investments in AI. This report, combined with the warning from OpenAI CEO about an AI bubble, has shattered Wall Street's optimism and triggered a sharp sell-off in tech stocks.

On Tuesday, the U.S. stock market's tech sector faced a significant setback, with the tech-heavy Nasdaq Composite Index falling 1.4%, marking its largest single-day drop since August 1. The blue-chip S&P 500 Index also closed down 0.7%.

The core beneficiary of the AI boom, NVIDIA, dropped 3.5%, while software company Palantir and chip design company Arm plummeted 9.4% and 5%, respectively. Meanwhile, defensive sectors such as consumer staples, utilities, and real estate rose against the trend, highlighting a clear shift of funds away from high-risk, high-momentum tech stocks.

The direct trigger for the market turmoil points to a key report released by MIT on Monday. The report states that as many as "95% of companies are seeing zero returns from their generative AI investments." This conclusion, combined with OpenAI CEO Sam Altman's recent comments about "investors being overly excited" potentially forming an AI bubble, severely undermined market confidence.

This sell-off concentrated on the best-performing stocks of the year, prompting investors to reassess the core logic supporting this round of gains—whether AI can truly and rapidly translate into corporate profits.

The "AI Gap" Revealed by the Report

The report, which traders described as "scaring people," is titled "The Generative AI Gap: The State of Business AI in 2025," published by MIT's NANDA project. The core finding of the report is that, despite high expectations for generative AI, the vast majority of projects have failed to produce tangible financial impacts.

A trader close to a multi-billion-dollar U.S. tech fund stated, "This story is causing panic among people."

The report is based on interviews with 150 corporate leaders, a survey of 350 employees, and an analysis of 300 publicly available AI deployment cases. The results show that only about 5% of AI pilot projects achieved rapid revenue growth, while the vast majority stagnated, having no measurable impact on the company's profit and loss (P&L) statement.

Aditya Challapally, the report's lead author, explained that the core issue lies not in the quality of the AI models themselves, but in the "learning gaps" of internal tools and organizations, as well as flawed integration strategies. Many executives attribute failures to regulation or model performance, but MIT's research points directly to issues in the corporate integration process.

For example, general-purpose tools designed for individuals, like ChatGPT, are popular due to their flexibility, but often hit roadblocks in corporate environments because they cannot learn or adapt from specific workflows

The Successful Minority and the Failing Majority

The report further analyzes the key differences between successful and failed AI deployments. The successful minority, such as some startups, have a strategy of "choosing a pain point, executing excellently, and establishing smart partnerships with companies using their tools." Aditya Challapally cites examples of some youth-led startups that "saw their revenue leap from zero to $20 million within a year."

However, for 95% of the companies in the dataset, AI implementation has not been effective. The report found that over half of the generative AI budget was allocated to sales and marketing tools, but the highest return on investment (ROI) actually came from back-office automation, such as cutting business process outsourcing and external agency costs.

Another key finding is that "buying" is better than "building." The success rate of purchasing AI tools from professional vendors and establishing partnerships is about 67%, while the success rate of building systems in-house is only one-third. Aditya Challapally points out that although many companies tend to build their proprietary tools, the data shows that this "lone wolf" model has a much higher failure rate.

This finding poses a direct challenge to companies that have invested heavily in trying to establish proprietary AI systems and raises questions about the capital expenditure efficiency of these companies. The report also notes that key success factors include empowering frontline managers rather than a central AI lab to drive applications, as well as choosing tools that can deeply integrate and adapt over time.

Valuation Pressure and Market Sentiment Shift

The emergence of this report coincides with growing concerns in the market about the high valuations of tech stocks. According to Bloomberg data, the expected price-to-earnings ratio of the Nasdaq 100 index is 27 times, nearly one-third higher than its long-term average. This stretched valuation sets the stage for a market correction.

Sam Altman's warning further fueled the fire. He candidly stated, "I do think some investors may lose a lot of money," and acknowledged that the market can experience "periods of irrational exuberance," which further exacerbated investor concerns.

Jacob Sonnenberg, a portfolio manager at Irving Investors, stated, "The market has been very hot, but today you see funds rotating out of many very hot, high-momentum stocks."

In fact, the market is highly sensitive to any negative news regarding AI.

In January of this year, DeepSeek announced a technological breakthrough with computing power far below that of its American competitors, which once triggered market turmoil. Although the stock price rebounded afterward, this incident highlighted the fragility of investor sentiment.

Tuesday's sell-off again proved that after months of AI hype, any evidence questioning its commercialization capabilities could serve as a catalyst for a market correction.

Risk Warning and Disclaimer The market has risks, and investment requires caution. This article does not constitute personal investment advice and does not take into account the specific investment goals, financial situation, or needs of individual users. Users should consider whether any opinions, views, or conclusions in this article align with their specific circumstances. Investment based on this is at one's own risk