POV: AI could increase corporate profits by $4.4 trillion a year, according to new research

Generative AI technologies like ChatGPT have taken the world by storm, thanks to their stunning ability to parse natural language and make decisions that are remarkably human. This latest advance in artificial intelligence is speedily transforming business and work, leaving companies, governments, and individuals scrambling to assess its impact in real time.

Our latest research on generative AI and productivity from the McKinsey Global Institute finds that generative AI has the potential to generate value equivalent to $2.6 trillion to $4.4 trillion in global corporate profits annually. There were 63 use cases in which we estimate that generative AI will raise productivity, including providing support interactions with customers, generating creative content for marketing and sales, and drafting software code based on natural-language prompts, among many other tasks. That would increase the value of productivity from artificial intelligence and analytics by 15% to 40% compared to previous generations of the technology—an amount that would roughly double as generative AI spreads more diffusely across the global workplace.

We estimate that generative AI is likely to deliver its biggest impact in banking, high tech, and life sciences as a percent of overall industry revenue. In banking, for example, the technology could create value equal to an additional $200 to $340 billion a year if all use cases were implemented. That’s not to say other industries won’t realize big value from deploying generative AI. All told, for instance, retail and consumer packaged goods companies could see an additional $400 billion to $660 billion in operating profits annually from the use of generative AI.

Across all industries, about three-quarters of the value from generative AI would emerge from four areas of business: customer operations, marketing and sales, software engineering, and research and development. In some use cases we studied, the technology would operate as a virtual expert, delivering information and making suggestions to, say, answer questions from a retail customer or develop creative content for marketing purposes. In other cases, the technology might be a virtual collaborator, creating candidate molecules for teams working to develop new drugs or suggesting and tidying code for a software engineer.

Much of the value of generative AI will come from growth in productivity across the economy—so long as employees affected by the technology shift to new work activities. We estimate that generative AI could increase labor productivity by 0.1% to 0.6% annually through 2040. Combined with other technologies, work automation could add an extra 0.2 to 3.3 percentage points to productivity growth.

Together with other technologies, the current capabilities of generative AI have the theoretical potential to automate work activities, such as drafting emails or responding to customer inquiries, that currently occupy 60% to 70% of the time employees spend working. In contrast, we previously estimated in 2017 that technologies available at that time could automate activities that occupied half of their time.

This is largely due to generative AI’s facility with natural language, which is required for work activities that account for about one quarter of total work time and beyond the capabilities of previous generations of artificial intelligence. Generative AI’s ability to converse with a user and draw conclusions from that conversation also means that it will have a greater impact on the knowledge work that plays a bigger role in occupations that have higher wages and educational requirements. That’s a marked difference from the effect of older artificial intelligence, which largely automated manual tasks.

Additionally, the pace of workforce transformation is likely to accelerate as a result of the higher potential for technical automation. We have developed scenarios to estimate the pace of actual adoption of these technologies across the global workforce, and, updated to account for the effects of generative AI, those scenarios now suggest that half of today’s work activities could be automated between 2030 and 2060. The midpoint falls in 2045, or about a decade earlier than our 2017 estimates. And while adoption might take decades across the global workforce, individual companies could transform their operations using generative AI much more quickly for a competitive advantage.

Employees will need support in learning new skills that help them work together with generative AI and, in some cases, in changing roles. Corporate leaders can begin by reexamining core business processes, and considering where best to apply generative AI and how employees can work with it, as well as determining what new skills and capabilities they may need. Only by actively managing the transition to generative AI—as well as evaluating the risks—can the significant economic promise of generative AI be fully realized.

Michael Chui is a partner at the McKinsey Global Institute. Lareina Yee is a senior partner and chair of the McKinsey Technology Council.