From six hours to 60 minutes: AI agent accelerates analysis work

• Ask one question in the chat instead of searching through scattered data silos
• Agents generate automatic analyses and charts within seconds
• Sovereign infrastructure delivers full AI power and the highest level of data security

Companies store enormous amounts of data, yet much of its potential often remains unused. Information is spread across dozens of databases and applications in different formats. Studies by McKinsey and the Microsoft Work Trend Index show that employees spend around a quarter of their working day simply searching for information. T-Systems addresses this data labyrinth: the platform “Talk to your data” searches, links, analyzes, and visualizes data with the help of an AI agent. Corporate knowledge becomes instantly accessible through a simple chat interface. Initial pilot projects already demonstrate significant efficiency gains. The time required for complex analyses has been reduced from six hours to less than 60 minutes.

Knowledge in dialogue: AI agent replaces time-consuming research

“Organizations do not suffer from a lack of data, but from the fact that their information is stored in separate systems,” says Sven Giesselbach, AI expert at T-Systems. “With ‘Talk to your data’, we connect these data sources and make a company’s knowledge directly accessible via chat.” Instead of searching through databases, spreadsheets, and documents one by one, employees can simply ask their questions in natural language. The AI agent analyzes the request, automatically searches the relevant systems, and delivers a structured and traceable answer within seconds, optionally including charts or visualizations. This fundamentally changes how companies access their knowledge. What once required time-consuming research becomes a direct dialogue with the available data.

Map of corporate knowledge

At the core of the solution is an ontology layer, a semantic layer that connects different data sources. It creates shared meanings between pieces of information even when they are labeled differently across various systems. “Together with our customers, we build this ontology and link the relevant data sources,” explains Giesselbach. “This creates a kind of map of a company’s knowledge, enabling AI agents to find the relevant information, combine it, and present it in a clear and understandable way.”

Real-world testing in hospital environments

The benefits become particularly clear in complex environments such as hospitals. Diagnoses, lab results, medication plans, and medical reports are often stored in different systems. With “Talk to your data”, doctors can retrieve this information directly. A single query is enough to access all diagnoses of a patient, for example, or to display the development of vital signs over several months in a chart. 
The AI agent also analyzes relationships between data sources and can point out possible inconsistencies, for instance when a laboratory value does not match the documented therapy. “The agent automatically sends multiple queries to different systems, evaluates the results, and checks whether additional sources might be relevant,” says Giesselbach. “This allows us to achieve a level of accuracy that can be reliably used even in sensitive medical environments.”

Measurable efficiency gains

Initial applications already show clear productivity improvements. In one project in the public sector, where extensive reports had to be researched and prepared, the required time dropped significantly. “What previously took around six hours can now be completed in less than one hour with the intelligent assistant,” Giesselbach reports. In addition to saving time, organizations benefit from the fact that information from multiple sources can be consistently combined through an intuitive, dialogue-based interaction.

Sovereign infrastructure for the next generation of AI

“Talk to your data” runs on T-Systems’ Agentic AI platform and uses its AI Foundation Services. These provide models, infrastructure, and security mechanisms. Building on this, “Talk to your data” enables knowledge management through a chat-based interface.

Data sovereignty is a central element of the solution. T-Systems operates “Talk to your data” entirely within its own cloud infrastructure, ensuring that sensitive information never leaves the protected environment. “Companies can harness the capabilities of modern AI without compromising on security or data protection,” emphasizes Giesselbach. The required scalability is provided by the Industrial AI Cloud, which supports both large language models and more specialized, efficient models.

That this approach reflects the broader market trend is confirmed by recent analyses from Deloitte, Gartner, and CIO.com. By the end of 2026, around 40 percent of all enterprise applications are expected to use AI agents. Specialized models, in particular, are projected to be used three times more frequently than large pre-trained language models such as GPT-4 by 2027.