Artificial intelligence has become one of the biggest trends in the corporate world. Companies across industries are investing heavily in AI tools, hoping to reduce costs, improve efficiency, and increase profits. However, a new study suggests that the real impact of AI in the workplace may be far smaller than expected.
According to research from the National Bureau of Economic Research (NBER), more than 90 percent of companies say AI has made no noticeable difference to employment levels or productivity over the past three years.
The study surveyed nearly 6,000 senior executives from the United States, the United Kingdom, Germany, and Australia. While two-thirds of executives said their companies use some form of AI, actual usage remains limited. On average, AI tools are used for only about 1.5 hours per week. Around 25 percent of respondents said they do not use AI at all in their workplaces.
When asked about employment impact, more than 90 percent of managers reported no change. Similarly, 89 percent said AI had not improved labour productivity, which was measured as sales per employee. In simple terms, despite the excitement around AI, most companies are not yet seeing higher output or major workforce reductions.
The most common uses of AI today include text generation through large language models, visual content creation, and data processing using machine learning. Tools such as ChatGPT and Microsoft Copilot are among the most widely adopted systems. However, large-scale financial gains remain rare.
Despite the limited short-term impact, business leaders remain optimistic. Executives expect AI to increase productivity by 1.4 percent and raise output by 0.8 percent over the next three years. About 75 percent of companies believe they will be using some form of AI by 2028.
At the same time, there are concerns about future job losses. The study estimates that AI could lead to around 1.75 million fewer jobs across the four countries by 2028, based on executive expectations.
This is not the first report to question AI’s business value. A separate study from the Massachusetts Institute of Technology last year found that 95 percent of organisations investing in generative AI were seeing zero return on investment.
The message from these findings is clear: while AI holds long-term promise, its immediate economic impact remains modest. For now, the technology appears to be more about potential than measurable performance.
