A recent Morgan Stanley study of British companies that have been using artificial intelligence in their processes for at least a year found an average productivity increase of 11.5% (with similar levels in the United States). Other recent (though non-academic) studies by OpenAI and Anthropic suggest that AI tools, such as ChatGPT, enable employees to save 40 to 60 minutes per day on work tasks, with 75% of employees reporting an improvement in the speed or quality of their work.
These results confirm an underlying trend: generative AI is no longer limited to one-off experiments, but is gradually establishing itself as an everyday work tool. And to understand what really explains these productivity gains, we need to look not at the technology itself, but at the concrete impact it has on the organization of work.
A reduction in cognitive load
In many organizations, a significant portion of working time is taken up by tasks that are not very visible but are time-consuming: sorting emails, searching for information, consolidating data, checking documents, rephrasing content, and preparing summaries.
This is precisely where AI is currently having the most tangible impact. Not by doing the work instead, but by lightening the load on teams. By summarizing, classifying, extracting, and suggesting, AI acts as an assistant that reduces the workload and allows people to focus on the essentials. This results in fewer interruptions and greater continuity in the work. Productivity does not necessarily increase because teams work faster, but because they work better, without unnecessary burdens.
Roles that evolve rather than disappear
Contrary to popular belief, AI rarely replaces a role as a whole, but rather transforms the composition of the tasks that make it up. Repetitive activities tend to decrease, while others become more important:
- validation of results
- arbitration
- understanding of processes
- human judgment on complex, ambiguous, or unexpected situations
In other words, AI shifts the center of gravity of work. Teams spend less time executing and more time supervising, interpreting, and deciding. The risk is therefore not the disappearance of jobs as such, but the gap between evolving roles and organizations that do not clearly redefine them.
Where does resistance come from?
In companies, resistance to AI is not limited to the fear of losing one’s job. It is often fueled by a lack of understanding: why is this tool being introduced, for what purpose, with what rules, and for what specific uses? Without clear communication, teams may feel that they are losing control, that their expertise is less recognized, or that they have to adapt to new technology on their own.
That’s why the most effective approaches are based on two pillars: transparent communication about the objectives and expected impacts, and progressive training that allows everyone to learn how to use the tool, gain confidence, and derive real value from it in their work.
AI as an organizational lever
AI does not create value on its own: it amplifies what already exists. In a structured organization, it streamlines processes, improves the quality of work, and strengthens teams’ ability to focus on important decisions. In a vague organization, on the contrary, it can accentuate irritants and areas of uncertainty.
According to a report by the World Economic Forum, nearly 23% of jobs worldwide are expected to change by 2027, due to the combined effects of automation and artificial intelligence. The report anticipates both a reduction in certain repetitive tasks and the emergence of new roles requiring different skills, particularly in analysis, automated system supervision, and decision-making.
But this evolution can be a major opportunity: AI paves the way for more efficient, more agile organizations and enriched professions, where humans refocus on added value—understanding, judgment, arbitration, relationships, and creativity. The companies that will benefit most will not be those that seek only to automate, but those that know how to support this change: clarifying uses, training teams, and methodically evolving roles. AI then becomes not only a lever for productivity, but also an accelerator of positive and sustainable transformation.
At TechNuCom, we support companies with a pragmatic “small steps” approach to quickly identify use cases that are easy to implement and achieve the fastest return on investment, while ensuring long-term buy-in from teams.
