MIT Report 2.0? Apollo: The Adoption Rate of AI by Large Enterprises is Declining

Recent studies have sounded the alarm simultaneously: enterprise-level artificial intelligence (AI) applications are facing severe challenges.

Three weeks ago, a study released by the Massachusetts Institute of Technology (MIT) stated that up to 95% of enterprises are receiving zero returns from their investments in generative AI. Last Sunday, Torsten Sløk, chief economist at Apollo Global Management, pointed out that the AI adoption rate among large U.S. companies is showing a downward trend.

The two studies reveal significant obstacles in the transition of AI technology from hype to practical application. Sløk cited official U.S. survey data indicating that companies with more than 250 employees are slowing down their AI adoption. This may signal a reassessment of the actual value of AI technology by enterprises.

The MIT report further analyzed the reasons behind this phenomenon, indicating that the issue lies not with the AI models themselves, but with flaws in the internal integration strategies of companies. The findings from these two studies have triggered a strong market reaction, leading to the Nasdaq index recording its largest single-day drop since August 1, with AI-related stocks like NVIDIA experiencing significant sell-offs.

Decline in AI Adoption Rate Among Large Enterprises

Torsten Sløk's analysis is based on a large-scale biweekly survey conducted by the U.S. Census Bureau under the Department of Commerce, which covers 1.2 million businesses and asks whether they have used AI tools such as machine learning, natural language processing, virtual agents, or voice recognition to help produce goods or provide services in the past two weeks.

The above chart presents the moving averages from six surveys conducted by the U.S. Census Bureau. The survey data shows that the AI adoption rate among large enterprises employing more than 250 people is declining. This trend suggests that while the market is enthusiastic about AI, large enterprises may be experiencing a "technology disillusionment phase" at the practical application level, beginning to reassess the actual value and return on investment of AI tools.

This decline in adoption rates may reflect the integration challenges faced by enterprises after initial attempts, as well as the difficulties in translating AI tools into actual business value. For investors, this data signal suggests that the commercialization path of AI technology may be more tortuous than previously expected.

MIT Study Reveals AI Investment Dilemma

The MIT NANDA project's report titled "The Generative AI Gap: The State of Commercial AI in 2025," released on August 18, provides a deeper analysis. Based on interviews with 150 business leaders, surveys of 350 employees, and analysis of 300 publicly deployed AI cases, the study found that only about 5% of AI pilot projects achieved rapid revenue growth.

The report's lead author, Aditya Challapally, pointed out that the core issue lies in the "learning gap" within enterprises and flaws in integration strategies. Many business leaders mistakenly attribute failures to the regulatory environment or model performance, overlooking the adaptation and integration issues within their organizations For example, general-purpose tools designed for individual users, like ChatGPT, are popular due to their flexibility, but often perform poorly in corporate environments because they cannot effectively learn from specific workflows or adapt to the specific needs of businesses. This "one-size-fits-all" application approach has led to many AI projects failing to deliver measurable financial impact for companies.

Key Differences in Successful AI Implementation

MIT's research also delves into the key differences between successful and failed AI deployments. A few successful companies, particularly some startups, have adopted a strategy of "focusing on a single pain point, executing precisely, and building intelligent partnerships." Challapally mentioned that certain startups led by young people have seen their "revenue leap from zero to $20 million within a year" using this approach.

The research found that over half of the generative AI budgets are allocated to sales and marketing tools; however, the largest return on investment actually comes from back-office automation, such as applications that reduce business process outsourcing and external agency costs. This indicates that companies may be misjudging the direction of their AI investments.

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 companies building systems internally is only one-third. This data poses a direct challenge to those companies investing heavily in attempts to create proprietary AI systems.

Market Reaction and Investment Impact

The findings from MIT had a significant impact on the market last month. The day after the report was released on August 20, U.S. tech stocks plummeted, with the Nasdaq Composite Index falling 1.4%. NVIDIA, a core beneficiary of the AI boom, dropped 3.5%, while Palantir and Arm fell 9.4% and 5%, respectively.

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

This shift in sentiment resonates with recent warnings from OpenAI CEO Sam Altman about "investors being overly excited," which may be forming an AI bubble, further intensifying market skepticism regarding the commercialization prospects of AI technology.

The release of the MIT report coincided with growing concerns about the overvaluation of tech stocks, with the expected price-to-earnings ratio of the Nasdaq 100 index at 27 times, nearly one-third higher than its long-term average.

For investors, these two studies provide important risk signals, indicating the need for careful assessment of the valuations and actual implementation capabilities of AI-related companies, rather than merely focusing on technological breakthroughs and market hype. The AI revolution may still be ongoing, but its path to commercialization is more complex and lengthy than anticipated