The economic effects of generative artificial intelligence

Generative artificial intelligence, particularly large language models (LLMs), is one of the major technological developments of recent years. This transformation raises major questions regarding its economic impacts, particularly on employment, productivity, and labor market dynamics.

The adoption of these technologies within American companies is growing rapidly. According to recent surveys, approximately 43,2% of American workers were already using generative AI tools by spring 2025, compared to only 30,1% a few months earlier, at the end of 2024. This rapid adoption phenomenon reflects both a general enthusiasm and a growing awareness of the operational benefits these tools bring. Generative AI is primarily used by young, well-educated workers in high-income positions, particularly in the information services, management, marketing, and information technology sectors.

The gradual widespread use of generative AI can be explained by its proven ability to significantly improve work productivity. Recent experimental studies reveal that the use of tools like ChatGPT or Gemini can significantly reduce the time required to complete specific tasks. For example, consultants using these tools can complete more tasks, faster, and with higher quality than those without access to these tools. On average, productivity is multiplied by three for tasks incorporating these technologies, allowing for substantial savings in working time. A task that usually takes 90 minutes could now be completed in just 30 minutes thanks to generative AI.

Despite this notable improvement in productivity, the intensity of use remains relatively moderate: only a third of regular users report using these tools daily. This means that generative AI is often used as a one-off supplemental tool rather than consistently throughout the day. Most users report less than 15 hours of use per week.

At the sectoral level, the impacts of generative AI are particularly noticeable in fields where repetitive, creative, or information-intensive tasks are common. For example, sectors such as customer service, marketing, and computer programming benefit greatly from these tools. Conversely, sectors such as agriculture, mining, and the public sector show much lower usage rates, likely reflecting the more physically or regulatory-sensitive nature of these jobs.

The overall economic impact of generative AI, however, remains to be determined. Some optimistic scenarios predict potential global GDP gains of around 7% within a decade, with a major job reorganization potentially involving the automation of up to 300 million positions. Conversely, cautious scenarios anticipate more modest gains and highlight increased risks of inequality and disruption in the labor market. Indeed, while some low-skilled jobs could be threatened by full or partial automation, highly skilled jobs or those requiring advanced analytical skills could see their productivity greatly amplified.

Furthermore, the use of these tools appears to offer significant benefits for job seekers. More than 50% of unemployed people who have used generative AI in their job search report finding new opportunities more effectively, mainly through assisted writing of resumes and cover letters, or preparation for job interviews.

However, the large-scale integration of generative AI also raises significant regulatory and governance challenges. The question of possible regulation of the data used to train these models, copyright, and associated ethical issues is becoming unavoidable. Some economists even suggest the creation of a public agency specifically dedicated to regulating artificial intelligence, in order to oversee its use in a way that maximizes its benefits while limiting societal risks.

Finally, it is important for policymakers, businesses, and educational institutions to anticipate and support these major transformations. The development of tailored educational programs and professional training geared toward the skills needed to effectively leverage these tools seems essential. Increased public investment in generative AI research and development could also promote a better distribution of the economic and social benefits of this revolutionary technology.

In short, generative artificial intelligence is profoundly changing the current economic and professional landscape. While the potential benefits in terms of productivity and efficiency are immense, the associated risks cannot be ignored. The major challenge in the coming years will therefore be to find a balance that maximizes these benefits while protecting the stability and fairness of the labor market. This evolution will require ongoing dialogue between businesses, researchers, regulators, and citizens to collectively define the contours of a digital future that benefits everyone.

To learn more: discover how the rapid rise of artificial intelligence redesigns the world, what we are taught AI and industrial revolutions et Why banning AI is not the solution.