The business context, data and AI architecture and governance, trends for companies in 2026, according to SDG

From the evolution of generative artificial intelligence (AI) and intelligent assistants through the business context, to the governancedata migrations and IA vertical by industries, SDG Group has presented the main trends that will define the future of companies in the coming year.

The consulting firm specialized in AI and Data Science has released its report in a meeting with the press this Wednesday ‘Data, Analytics & AI 2026’which highlights ten technological innovations and business that will define competitiveness in organizations.

These trends are reflected in the SDG Innovation Radaryour periodic and dynamic analysis tool on the technology sector to identify the main emerging trends. Specifically, the company analyzes four times a year the evolution of the different tools, methodologies and emerging architectures in the areas of AI, data technologies, data architecture and AI and their impact on the business.

It should be noted that, for the first time, this radar is no longer an internal SDG tool and becomes “a living platform” that is now available so that any organization can consult it on the company’s website.

TRENDS FOR ARTIFICIAL INTELLIGENCE

In this sense, the first quadrant of the Innovation Radar encompasses the trends entirely related to AI, taking into account a current framework in which the important thing is not simply to implement this type of technology in the company, but to know how to do it so that it becomes a real competitive advantage.

In this regard, as the head of Innovation at SDG Group, Miguel Romero, has explained, a key point is the need to add new semantic layers to generative AI that allow companies to carry out their own data analysis service to obtain ‘insights’.

This means that the large language models (LLMs) that drive generative AI need a solid semantic layer, clear business rules, and data dictionaries that allow them to understand structures, because “AI needs meaning, not just data.”

Linked to this point, the report also includes how model manufacturers are expanding the context windows of AI models, developing specialized ‘software’ that facilitates the creation of agent memory, the monitoring of user conversations or the implementation of guardrails.

“The real challenge for 2026 is to propose a new basis in which data provides and understands context and meaning so that AI systems understand the business,” Romero stated.

Likewise, it also points to the acceleration of foundational AI modelswhich not only promote text generation but also other techniques in the world of machine learning.

MIGRATIONS AND METADATA: PRIORITY DATA TECHNOLOGIES IN 2026

Regarding the second quadrant of the radar, which collects the data technology trends that are seen as priorities this year, SDG has highlighted the importance of migrations as strategic initiatives.

Specifically, Romero has referred to the use of Generative AI as a method to transform system migration processesturning them into more efficient strategic projects, instead of the processes based on risky decisions and high costs that they currently entail.

This will allow companies to achieve a competitive advantage because they will be able to carry out faster and easier data migrationsensuring that there is no fault, for example, when modernize your systems.

He also made reference to the metadata operating systems. That is, active, AI-driven metadata structures that transform this information into a layer of operational intelligence. As a result, structured or unstructured data is reliably traceable and machine processable.

By allowing advanced management of metadata, technical information is simplified, making it more accessible so that users or companies can “understand, organize and leverage your data.

NEXT GENERATION ASSISTANTS AND AI GOVERNANCE

On the other hand, with respect to the data architecture and AIthe SDG Innovation Radar has pointed to new generation assistants as one of the most relevant trends.

Romero has referred to how, during the past year, the use of assistants was relegated to a question and answer interaction. However, during 2026, the trend will be to move towards use of new AI assistants that improve access to internal information in a more “natural and efficient” way.

These new assistants will offer more complete answers and, therefore, more useful to improve decision making, since they allow more interactive process in which the AI ​​reasons jointly with the user.

Following this line, given the rise in the implementation of AI in companies, another relevant point will be the observability and governance of this technology. That is, 2026 will be a year in which we commit to incorporating supervision and management capabilities that guarantee reliable use of the systems in a manner aligned with the company.

Romero has pointed out that, in the case of the European Union, this trend is “almost mandatory” and is related to the implementation of permissions and surveillance methods on the devices, assistants and agents used in a company.

Likewise, the AI on edge devices is another relevant factor, which refers to how new specialized ‘hardware’ allows AI solutions to be run directly on devices and offers new use cases with “more efficient and resilient” systems, while at the same time improves response speed and data privacy.

This ‘hardware’ is useful in cases where there is no stable connection to use the cloud reliably or where they cannot depend on the connection for their operation. For example, factories that are in somewhat isolated environments, for autonomous driving or robotics.

VERTICAL AI AND THE RISE OF PROBABILISTIC MODELS

Finally, in the business quadrant and other sectors, SDG has highlighted the Vertical AI by industries. That is, AI solutions designed specifically for specific sectors, so that companies can adopt this technology for your needs more effectively.

This shift from offering generalist AI to solutions with more specific knowledge is seen in providers such as Anthropic with its Claude AI, which later went on to offer the specific Claude for Financial Services service.

The latest trend included in the report is the rise of probabilistic models, instead of deterministic models. This means “breaking the pattern of solving each variant through explicit programming” and moving on to automating problem resolution.

GROWTH OF SDG GROUP IN 2025

Aside from the new trends for the sector, Romero has also reviewed the company’s progress during the past year 2025, with a 12 percent billing growth, reaching the 120 million euros in Spain and 202 million euros globally.

Likewise, the SDG Group team in Spain has grown to have 1,700 employees Currently, together with the rest of the global teams, they provide service to more than 300 active clientsof which 80 percent are implemented in the IBEX 35.

On the other hand, Romero has stressed that they have grown by 50 percent in its AI services, given that these types of key technologies for companies are becoming “more and more democratized” and, therefore, their implementation in companies is increasing.