- 96% of European companies have or will use AI agents
- Users of AI agents are more likely to be optimistic than non-users
- Data silos and poor data integration between systems are a main obstacle
Artificial intelligence has become so deeply rooted in our life now that companies are looking to go further by adopting even more AI agents in the months and years to come, said research.
The new Mulesoft data show that almost all European companies (96%) have implemented or implemented AI agents – defined as partly autonomous AI tools which can perform tasks, make decisions and interact With users – by 2026.
The news comes as companies expect IT workloads to increase 18% between 2024 and 2025, AI agents promising to offer tangible advantages of efficiency.
The AI agent era is on us
More than nine in 10 computer leaders (93%) think that AI will increase the productivity of developers, but this number is even higher (98%) among users of AI agents, which indicates that agents of Support has provided so far and return on investment believe that they “I take away from technology.”
“To unlock this new capacity, integration and APIs are essential to the construction of a ready -made foundation for the agent. Agents depend on unified data and business actions, supplied by integration and automation, to provide precise and reliable responses and perform complex and several steps, “noted Mulesoft SVP and GM Andrew Comstock .
As competent as they are, the adoption of AI agents is always with its challenges. More than two thirds (69%) of European companies cite the integration of data as a major challenger for the adoption of AI, and 94% of survey respondents have trouble integrating data between systems . In addition, only one out of three application (32%) is generally connected, which makes agents of IA ineffective in many cases.
Ninety percent of companies agree that data silos create commercial challenges, but those who try to use AI agents more in agreement (94%).
It is therefore clear that a solid data foundation that is ready for the agent is necessary before companies even plan to develop their AI strategies.
