Artificial intelligence, or AI, has become the main axis of transformation within the modern workplace. Since 2023, with the massive emergence of generative models, AI-driven automation tools, and AI-based intelligent assistants, companies have experienced an accelerated revolution that has changed the way people perform their daily tasks.
Today, in 2025, it is common for workers in virtually all sectors to use AI systems to write emails, generate documents, synthesize information, program, analyze data, and automate repetitive tasks. The impact of AI on individual productivity is undeniable. However, a recent study disseminated by international media and conducted by Atlassian introduces a crucial nuance: personal efficiency driven by AI is increasing significantly, but at the same time, the structure of teamwork is deteriorating. What seems to be an indisputable advance thanks to AI carries a hidden cost that organizations are only beginning to recognize.
According to this study, employees who use AI tools and artificial intelligence report an increase of around 33% in their daily productivity. This translates into an estimated savings of 1.3 hours per person per day, a considerable number if extrapolated across an entire organization. Nonetheless, despite this individual boost provided by AI, only a small percentage of companies have noticed significant improvements in their overall efficiency. The diagnosis is clear: personal productivity driven by AI does not always translate into organizational performance.
This article by ITD Consulting delves into this phenomenon, examines the cultural, labor, and psychological implications of widespread AI adoption, incorporates evidence from complementary studies, and offers an overview of the challenges companies face when seeking to integrate AI without compromising team cohesion.
The Initial Promise: Efficiency, Speed, and More Time to Think
When the first generative AI tools appeared in the workplace, the main goal was clear: to free employees from repetitive work through AI and allow them to focus on higher-value strategic tasks. The benefits of AI were evident from the start. Many workers were able to accelerate report writing with AI, obtain automatic meeting summaries generated by AI, create presentations in minutes, and automate processes that previously took hours thanks to AI systems. For the first time, AI-based technology not only helped execute tasks but also collaborated in the creation of ideas and content.
Data collected by Atlassian shows that this enthusiasm for AI has a real basis. Employees using AI not only work faster but also feel more confident when making operational decisions assisted by AI. Additionally, a significant portion of them prefer to consult an AI system first rather than a colleague when facing a question. This reveals a profound change in the way people seek information and solve problems: AI has become an everyday interlocutor, available immediately, without judgment, without hierarchies, and capable of responding in seconds.
Even business leaders have quickly adopted these AI tools. A significant proportion of employees report having seen their bosses use AI applications in real time. This detail reflects that AI adoption is not limited to operational levels but crosses the entire organizational structure. Furthermore, most employees perceive that their leaders openly encourage the use of new AI-based technologies and continuous experimentation, creating an atmosphere conducive to AI-driven innovation.
However, despite this enthusiasm for AI and the immediate benefits it provides, the story is not as linear as it seems. The question arises: if we are all more productive individually thanks to AI, why do we not see equivalent leaps in company-wide productivity?
The Hidden Cost: Deterioration of Collaboration and Increased Silos
The same study that celebrates the increase in individual productivity also warns of a troubling phenomenon: AI is contributing to weakening teamwork. Only a small group of organizations report having observed significant improvements in overall efficiency after implementing AI on a large scale. This suggests that the individual benefits provided by AI are not translating into collective gains.
One of the main reasons is fragmentation. As each worker or department adopts its own AI tools, the organization begins to divide into technological islands shaped by different types of AI. Each team has its own preferences, automated AI workflows, AI-generated templates, and particular dynamics. This generates a lack of coherence in processes that were previously shared and hinders cross-functional collaboration, because information no longer flows uniformly. Instead of breaking down barriers, AI can reinforce organizational silos.
Another problem is that decision-making becomes more unequal. Teams or individuals with greater mastery of AI technology gain disproportionate influence within the organization because they can produce more in less time and dominate the AI tools that generate key inputs. Meanwhile, those who are less familiar with AI fall behind. This creates internal tensions, inequalities in workloads, and even friction in the perception of individual value within the team, especially when AI amplifies performance differences.
Another concerning finding is that many business leaders have observed moments when AI, far from helping, introduced confusion or generated rework. In some cases, employees rely too heavily on results automatically generated by AI and do not verify them carefully enough, leading to errors that must be corrected afterward. This consumes time, generates frustration, and demonstrates that AI, if not properly supervised, can become an obstacle in tasks that require precision.
Finally, there is a direct impact on innovation. Companies that focus their AI strategy solely on improving individual productivity often report fewer advances in collective creativity and innovation. This occurs because innovation depends heavily on the exchange of ideas, interdisciplinary collaboration, and the social construction of knowledge. When each person works with their own AI digital assistant and consults colleagues less, that exchange is weakened, and joint creativity is reduced, showing that AI can strengthen the individual but weaken collective intelligence.
Additional Evidence: What Other Recent Studies Say
The results of the Atlassian study are not an isolated case. Several recent academic works are exploring how teams behave when AI enters their dynamics and how these AI tools affect human interaction. The expansion of AI in collaborative environments has created a new field of research that examines how teams adapt when living alongside AI systems in their daily activities.
One of these studies, conducted by researchers specializing in organizational behavior, demonstrated that the personality of the AI agent directly influences joint performance with humans. In some experiments, teams composed of a human and an AI with a particular “personality” produced higher-quality work, although in smaller quantity, while other combinations with different AI profiles favored quantitative productivity. This research shows that AI is not neutral: its design, communication style, and simulated behavior decisively influence team dynamics, either enhancing or limiting the effectiveness of collaboration with AI.
Another experiment analyzed teams using hybrid collaborative systems, integrating humans and artificial AI agents to design advertising campaigns. The results showed that human-AI teams significantly increased their level of communication compared to teams composed only of humans. Humans spent less time on manual editing, as AI took over those tasks, and more on conceptual content, freeing up creative space. However, an interesting finding emerged: in image generation tasks, purely human teams produced better visual results than hybrid human-AI teams, suggesting that AI does not enhance all human skills equally and that the human-AI combination may vary depending on the type of task.
There have also been recent advances in AI tools aimed at improving team cohesion. Some AI applications analyze internal conversations and provide automatic feedback on clarity, participation, or alignment with goals. In experimental studies, these AI systems enabled teams to work more harmoniously and with fewer misunderstandings. Another AI system, designed to detect misalignments in real time, offered minimal but effective interventions to redirect the conversation or correct the project’s course, demonstrating that AI can act as an active facilitator of group communication.
These complementary studies indicate that AI can be a powerful tool not only for accelerating tasks but also for facilitating collaboration, as long as it is well-designed and properly integrated into collective processes. The problem is not AI technology itself, but how it is used, under what conditions, and with what strategic objectives it is incorporated into human dynamics.
Invisible Consequences: Internal Inequality, Questionable Quality, and Social Isolation
In addition to structural effects on teams, widespread AI adoption generates a series of less visible but equally relevant consequences for the future of work. The accelerated expansion of AI not only transforms processes but also behaviors, perceptions, and internal dynamics within organizations.
One of these consequences is the growing gap between those who master AI and those who do not. In many companies, employees with advanced skills in AI tools already receive better evaluations, more opportunities, and, in some cases, higher salaries. Recent studies show that in certain sectors, the wage difference between those who manage AI and those who do not can exceed 50%. This AI-driven disparity creates internal tensions and generates a sense of inequality that can erode team cohesion, especially when AI becomes a strategic resource that not everyone has equal access to.
Another emerging phenomenon is the proliferation of what some experts call “AI junk content.” In certain work environments, the accelerated production of drafts, reports, or automatic documents generated by AI has created an abundance of low-quality information that must later be reviewed or corrected by humans. This rework not only consumes additional time but also deteriorates the perceived quality of AI-generated work. Some employees report considering colleagues who rely excessively on AI tools without adequate supervision as less competent, which can generate distrust, tensions, and significantly affect workplace relationships.
Additionally, there is an emotional and social impact associated with excessive AI use. By replacing many human interactions with AI consultations, the density of social networks within the organization is reduced. Some research has shown that when people have fewer opportunities to interact directly, discuss ideas, or solve problems together—activities that previously occurred without AI mediation—team psychological safety decreases, and feelings of isolation increase. This not only affects employee well-being but also harms the team’s ability to innovate, solve complex problems, and make well-informed collective decisions, demonstrating that AI, if not balanced with human interaction, can weaken internal social cohesion.
How to Prevent AI from Weakening the Team?
Although AI has a positive impact on individual performance, companies need clear strategies to prevent this AI-driven benefit from becoming a collateral damage for teamwork. Integrating AI into business dynamics requires a conscious and planned approach so that AI empowers teams rather than weakens them.
A first recommendation is to redefine success metrics. Productivity should not be measured only in terms of time saved thanks to AI or tasks completed with AI. Organizations need to incorporate metrics that value collaboration, collective innovation, and knowledge transfer between areas—goals that can also be reinforced by well-implemented AI systems. AI use should be linked to shared organizational objectives, not just the individual performance that AI facilitates.
A second fundamental step is to avoid technological fragmentation. The use of many AI platforms in isolation creates inequalities and hinders coordination. To prevent this, companies can opt for unified AI systems that integrate tools, promote collaboration, and facilitate the flow of information between departments. It is also recommended to create internal repositories with best practices, templates, examples, and knowledge generated from AI use, so that the entire team can benefit equitably and avoid AI deepening internal gaps.
Another key recommendation is to design or select AI agents appropriate to the organizational culture. Research on AI agent personality shows that the system’s behavior directly influences team performance. Therefore, organizations can experiment with different communication styles, configurations, and functionalities in their AI agents until they find which integrate best with human team dynamics and enhance collaboration rather than obstruct it.
It is also useful to implement AI systems that act as team coaches. These AI tools can provide clarity, help resolve misunderstandings, analyze communication patterns, and offer timely feedback. In this way, AI does not act only as a task executor but as a facilitator of interaction, cohesion, and strategic alignment.
Finally, it is essential to promote a co-intelligence culture. This involves training employees not only in the technical use of AI but also in how to make ethical decisions with AI, supervise work generated by intelligent AI systems, and interpret the results produced by AI. The goal is for humans and machines to work together complementarily, not competitively. AI should amplify human value, not replace it, and its implementation should strengthen collective intelligence rather than fragment it.
AI offers extraordinary opportunities to improve individual performance and optimize processes within organizations. However, AI’s impact on collective productivity is not automatic: it depends on how AI is integrated into team dynamics, how its use is supervised, and how aligned it is with organizational culture. When AI is applied strategically, it can enhance collaboration, improve knowledge transfer, and accelerate innovation, becoming a true driver of transformation. Conversely, if AI is implemented in isolation or without direction, it can fragment work, generate inequalities, and weaken team cohesion.
To fully harness AI’s potential, companies must create technological ecosystems that break down silos, encourage communication, and reinforce collective intelligence. This involves establishing metrics that value both individual productivity and group collaboration and training teams to work complementarily with AI. If you want to implement AI solutions that strengthen both efficiency and cohesion in your organization, the experts at ITD Consulting can help you design and integrate these tools strategically. Write to us at [email protected] and discover how AI can transform your company safely and effectively.
EN 
ES