AI Accelerates Employees, but Companies Struggle to Keep Up

AI and the Transformation of Organizational Structures

Artificial intelligence is profoundly changing the way information is produced and processed within companies. Historically, organizational structures were designed to manage the flow of information, a process that AI now accelerates, challenging these established structures. This transformation raises a crucial question: how can companies convert this new speed into a collective advantage?

A Paradox of Efficiency

AI does not directly destroy jobs, but it undermines existing organizational charts. For decades, companies have been built around the essential function of information flow, involving data collection, synthesis production, analysis consolidation, and decision preparation. Today, AI is capable of accelerating these processes, which mechanically weakens certain organizational structures inherited from a time when information was slower to produce and consolidate.

The Gap Between Individual Speed and Organizational Inertia

Over the past two years, AI tools have spread remarkably fast within organizations. They have become everyday tools for document writing, information synthesis, analysis preparation, and presentation structuring. The time required to accomplish these tasks has significantly decreased, with work that once took an hour now being completed in just a few minutes. At first glance, this promises a significant productivity gain.

However, empirical observations urge caution. A study by the consulting firm Asterès on the early uses of AI in artisanal businesses estimates the average gain at about 2.1 hours per week, which is just over 5% of working time. This progress, while real, remains modest on the scale of overall productivity.

Necessary but Insufficient Reorganization

EY's Work Reimagined Survey reveals another gap: although AI is widely used by employees, only 28% of organizations have reorganized their teams to fully exploit its potential. Furthermore, the Boston Consulting Group highlights that nearly half of users save more than an hour a day thanks to AI, but only a third receive guidance on how to use this saved time.

Challenges of Control and Coordination

The rapid spread of AI-generated content raises an age-old question in organizations: that of control. Who validates? According to what criteria? With what level of traceability? When these rules are not clarified, companies often compensate with informal loops of proofreading or coordination, which ultimately neutralizes some of the productivity gains.

An Economic Impact Still Diffuse

The initial waves of AI adoption often produce a paradoxical effect: much activity, but an economic impact that remains diffuse. Productivity gains related to AI often manifest as small individual improvements that do not always translate into a significant increase in overall productivity. Moreover, the acceleration of work can lead to an increase in demands and iterations, thereby diluting potential gains.

Towards Structural Transformation

To capitalize on these gains, some companies are beginning to adopt a more structural approach, carefully observing workflows and identifying points where AI can truly transform the economics of an activity. This approach aims to turn AI into a genuine competitive advantage rather than just a tool for individual enhancement.