With astonishing speed, agentic artificial intelligence has moved from being a futuristic horizon to a very concrete reality in the business environment. This technology differs from the AI assistants we knew until now in a key aspect: it can perform complex tasks autonomously, planning and executing multiple steps within a workflow, without direct human intervention.
A McKinsey study reveals strong curiosity about AI agents among organizations. Twenty-three percent of respondents stated that they are already scaling an agentic AI system in some part of their company, while an additional 39% indicated that they have begun experimenting with this technology. However, usage is not yet widespread nor fully extended at the operational level. Less than 10% reported that their organizations are scaling AI agents; among those that are, most are doing so in only one or two functions.
According to the same study, the use of AI agents is reported more frequently in IT and knowledge management areas, with use cases such as help desk management or advanced internal information research. By industry, adoption is more frequent in technology, media, and telecommunications, as well as in the healthcare sector.
What’s coming next: AI agent ecosystems
Although AI agents are not yet fully mature and continue to require supervision and governance, in the short term a scenario is anticipated in which they will begin to interconnect with one another. In this way, they will give rise to true ecosystems of autonomous agents capable of managing multiple operational circuits within organizations.
Gartner’s projections are emphatic: by the end of 2026, 40% of enterprise applications will incorporate task-specific AI agents, compared to less than 5% in 2025. In 2027, collaboration among AI agents within a single application will begin to be observed, and one third of implementations will combine agents with different skills to manage complex tasks across application and data environments. Meanwhile, starting in 2028, AI agent ecosystems will enable networks of specialized agents to collaborate dynamically across multiple applications and business functions.
In the medium term, Gartner provides another key forecast: by 2035, agentic AI could generate approximately 30% of enterprise application software revenue.
From this perspective, the impact of this technology will transform enterprise applications “from tools that support individual productivity into platforms that enable fluid autonomous collaboration and dynamic workflow orchestration.”
Without strategy, there is no paradise
For companies to fully capture the advantages of this new stage of automation promised by agentic AI, it will be essential to develop strategies aligned with their business objectives and operational capabilities.
According to IDC analysts, the effectiveness of these systems will depend “on how effectively organizations learn to design, implement, and manage them,” reinforcing the need to combine technological innovation with planning, governance, and the right talent.
In this context of accelerated transformation, understanding the main technology trends is key to anticipating investment decisions and digital evolution. For this reason, it is important to review other trends related to hybrid infrastructure, application platforms, and cybersecurity, which are compiled in the LOL Trends 2026 ebook, available for free download here.
