The lack of infrastructure represents a brake on the advancement of AI for 73% of IT managers in Spain

The main barrier to scaling artificial intelligence (AI) is not the lack of investment, but of infrastructure, as stated by 73 percent of IT managers in Spain, and to take the pilots to production, data streaming appears as the tool that allows having reliable and updated data.

Almost three-quarters (72%) of IT managers worldwide see their efforts slowed down by the lack of systems capable of processing data in real time, as concluded in the new ‘Data Streaming Report 2026’, from Confluent.

In Spain, the situation is similar: 73 percent point to this same lack as their main obstacle, and 71 percent acknowledge having faced at least three major challenges on the path to the adoption of AI.

The report, based on a survey of 4,625 IT managers around the world, analyzes the challenges that companies face in scaling artificial intelligence and points out that the key is to strengthen the infrastructure on which these initiatives are based, rather than increasing investment in AI itself.

According to the research, 72 percent of global IT leaders have encountered at least three challenges when scaling artificial intelligence initiatives: lack of infrastructure to process data in real time (72%), uncertainty about the lineage and quality of data (66%), and fragmentation in data management (65%).

The Spanish technology sector faces a very similar panorama: the lack of infrastructure for real-time data also leads its list of obstacles (73%), followed by the shortage of talent specialized in AI and data (67%), the fragmentation of data between systems (64%).

IT infrastructure challenges are also slowing the deployment of agentic AI. Two-thirds (66%) of global IT managers point to data infrastructure and information quality problems as the main barriers to its adoption, a figure that rises to 67 percent in Spain.

Both globally and in Spain, 32 percent claim to already have solutions deployed in production, while the majority continue to face delays.

UNLOCK AI IN REAL TIME

Making the leap from pilot projects to production increasingly depends on the data that underpins AI. In Spain, 79 percent of IT managers already consider the use of corporate data to drive their artificial intelligence systems a priority; The quality, timeliness and context of the data are as crucial to the success of AI as the model itself.

Faced with this challenge, data streaming appears as a widespread response among Spanish IT teams. Nine out of ten managers say that these platforms improve the quality, context and accessibility of information, making it a reliable asset for AI systems.

A trust that also translates into concrete expectations: in Spain, 92 percent expect data streaming to increase the impact of their investments in artificial intelligence, and 90 percent consider that it accelerates the path towards its adoption.

The report also points out that data streaming has overtaken AI itself on the priority list of Spanish IT managers. 87 percent of respondents place it as a strategic priority, above AI and machine learning solutions (84%),

This demonstrates a change in approach in the sector: before investing more in models, organizations want to ensure that the data that feeds them is reliable, up-to-date and arrives at the right time.

“Models must be connected to the systems, events and signals that reflect what is happening throughout the company. Companies that are making the most progress are not only investing in AI itself, but also in the databases necessary to support it. Those foundations will be what determine which organizations manage to convert investment in AI into large-scale business value,” added Confluent’s product director, Shaun Clowes.