Highlights 2025 – Artificial intelligence at Fraunhofer IIS

Artificial intelligence is a major research focus at Fraunhofer IIS. For many years, our teams have been developing AI systems designed for efficient use across a wide range of domains and industries.

We pursue two complementary approaches: with specialized, task-oriented AI, we deliver tailored, highly efficient solutions. At the same time, we drive the development of broadly applicable foundation models and methods – whether universal or optimized for specific domains.

The foundation models currently attracting major attention are large, potentially multimodal, universally pre-trained AI systems that provide the basis for rapidly deploying AI in specific applications. Among them is the Fraunhofer language model Teuken-7B, trained in 24 official European languages with exceptional resource efficiency. Through our work on Teuken, Fraunhofer has built the capabilities needed to process large volumes of data and to develop specialized foundation models. In doing so, we help safeguard the digital sovereignty of Germany and Europe.

Task-oriented AI refers to technologies developed to perform specific tasks. This includes both foundation models and foundational methods such as machine learning, deep neural networks, and forecasting or optimization techniques. These approaches are designed to deliver optimal solutions for our customers. Whether in data selection, the configuration of training procedures, or algorithm and hardware development, we conduct research on resource-efficient, cost-effective, low-energy, low-maintenance, time- and data-efficient systems with low latency for real-world domain applications.

Our project examples illustrate how task-oriented AI is applied in practice:

 

In signal processing, the institute has further strengthened its leading research position through the use of AI in audio and speech processing.

 

In the field of AI-based image analysis, we conduct research on medical and agricultural applications as well as on industrial inspection systems, including those used in waste-sorting facilities.

 

To reduce hesitation around integrating AI, our demonstration production line in Dresden offers companies the opportunity to test AI deployment in manufacturing.

 

In the Antrieb 4.0 project, we use state-of-the-art data analysis methods to enable fast and efficient access to heterogeneous data.

At the same time, we are contributing our many years of microelectronics expertise to the development of AI chips such as SENNA. Our goal is to implement AI far more efficiently and thereby enable significantly more resource- and energy-efficient deployment. Our vision is to make large models available for local use across all classes of devices in the future – from small hearing aids to large industrial machines.

SENNA-Chip

In-demand AI Expertise

We are a key partner in advancing innovative future technologies such as artificial intelligence. In 2025, we showcased our AI expertise at high-profile events, in expert committees, and through widely noted publications.

IHK Meets Science: Language Models »Made in Germany«”

At an event hosted by IHK Nürnberg für Mittelfranken (Nuremberg Chamber of Industry and Commerce) on February 17, Institute Director Prof. Bernhard Grill presented insights into the European AI language model Teuken-7B. The discussion that followed underscored the strong interest within industry in AI solutions »Made in Germany«.  

 

Strong Interest in the Publication

Our book »Unlocking Artifical Intelligence: From Theory to Applications« attracted considerable attention in 2025: the open-access volume has already been downloaded nearly 200,000 times. Around 70 experts contribute insights into key research and application areas of AI.

Workshop »Blue Swan AI Gigafactory«

On November 24, the event exploring a potential European high-performance computing and data center in Northern Bavaria brought together around 90 leading figures from science, industry, and public administration. Together, they advanced the vision of greater digital sovereignty and economic competitiveness.

Bavarian AI Council in Nuremberg

On July 1, the Bavarian AI Council held its summer meeting on our premises. In his opening remarks, Institute Director and council member Prof. Alexander Martin emphasized the importance of AI for the future of research and innovation.

 

Back to Annual Report 2025