A recent 69-page report from OpenAI, published on August 11, offers a complex view of artificial intelligence adoption within enterprises, revealing a significant disconnect between the perceived benefits and quantifiable financial returns. While the document broadly paints a picture of exponential AI usage growth across various seniority levels and job functions, a specific detail buried on page 35 challenges the direct correlation between AI use and a key financial metric: revenue per employee. The research indicates no statistically significant link between the volume of messages sent or tokens consumed by employees and a company’s revenue per employee, once other control factors are considered.
This finding introduces a layer of nuance to the narrative often presented by the rapidly expanding AI industry, which saw companies like the three-year-old Stockholm-based firm Lovable double its valuation to $13.3 billion recently. The report acknowledges that companies with higher existing revenue per employee tend to be early adopters of ChatGPT, and those that use the technology more also generally exhibit higher revenue per employee. However, this observation does not definitively establish that increased AI usage *causes* an increase in revenue per employee. Instead, it suggests that already lucrative companies are simply more inclined to integrate AI tools into their operations.
The report also highlights a surprising pattern in AI engagement across different organizational tiers. Executives, despite their critical role in strategic decision-making, appear to be among the least intensive users of AI tools. Data on page 29 shows that senior employees engage with these platforms less frequently, measured by weekly messages per user. Conversely, early career employees demonstrate the highest usage rates, a point emphasized by OpenAI CFO Sarah Friar in a LinkedIn post, suggesting that competitive advantage might increasingly stem from those “closest to the work.” This usage disparity raises questions about who within an organization is best positioned to assess the true return on investment for AI initiatives.
Further complicating the picture, OpenAI’s enterprise usage, specifically in terms of output token growth, experienced a notable plateau from October to December 2025. During this period, competitors like Anthropic’s Claude Code reportedly gained significant traction in the corporate sector. While the growth curve dramatically surged upward in January 2026, which OpenAI attributes to both new client acquisition and deeper engagement from existing customers, this earlier stagnation suggests that the path to widespread, consistent adoption is not always linear. Sam Altman, OpenAI’s CEO, has reportedly been restructuring the company to prioritize enterprise sales, even shelving internal projects like the video app Sora, underscoring the strategic importance of this market.
In a clear move to accelerate growth and potentially prepare for a future IPO, OpenAI recently appointed Dali Rajic as its new Chief Revenue Officer, replacing Denise Dresser, who held the position for less than a year. Rajic’s mandate includes boosting customer adoption and assisting businesses in quantifying the impact of AI, directly addressing the very ROI questions raised by the report itself. This aggressive hiring decision signals OpenAI’s commitment to translating its technological advancements into clear business value for its clients.
The credibility of such internal reports often comes under scrutiny, and OpenAI’s latest publication is no exception. While two of the five authors are academics affiliated with Columbia Business School and Wharton at the University of Pennsylvania, a footnote clarifies that they contributed to the work as paid contractors for OpenAI. This arrangement, while not uncommon, does muddy the waters regarding the absolute impartiality typically associated with independent academic research. Ultimately, the report serves as a reminder that while the AI industry continues to attract billions in investment and generate considerable excitement, companies exploring these technologies will need to rigorously measure their own impact, rather than solely relying on industry hype or even sponsored research.
