Non-scientific publications

Our non-scientific publications provide practical insights and reports on the work of the DLR Institute for AI Safety and Security.

June 2026, DLRmagazine

What causes it to go off course?

In order for autonomous vehicles to participate safely in road traffic, they must be able to reliably recognise road signs. To this end, they are fed large amounts of training data. However, if this data is manipulated – for example, by altering pixels in images – the driver assistance system can be misled. In the worst-case scenario, a stop sign could be interpreted as a 'give way' sign or a tampered sticker could be taken as a signal to 'go full throttle'. Researchers at the DLR Institute for AI Safety and Security are developing methods to protect the AI algorithms of such applications against such 'data poisoning' – to ensure that vehicles do not go down the wrong track.

June 2025, DLRmagazine

Base-X connects data for sustainable mobility

Base-X emerged from the Gaia X 4 Future Mobility project family. Over a period of three and a half years, 80 companies and research institutions worked on more than 20 use cases - all from the mobility sector. The DLR Institute for AI Safety and Security is responsible for coordinating the project family. Gaia-X is a European project and aims to create a powerful, secure and competitive data infrastructure. Base-X seamlessly connects existing data and service infrastructures with Gaia-X-compliant data spaces and data ecosystems.

November 2024, KölnBusiness Magazin

Insights into the DLR Institute for AI Safety and Security: From autonomous driving to aviation

The "Gaia-X 4 Future Mobility" project is being driven forward at DLR's Institute for AI Safety and Security under the leadership of Maximilian Stäbler. The idea: a digital ecosystem in which companies and local authorities can access AI-supported solutions for traffic control. DLR has been able to attract more than 80 partners for this project. German tech companies are also involved. One example of an application: Rubbish trucks equipped with cameras that detect potholes and analyse the data automatically.

April 2023, inware - The IAV magazine about software

AI - more science than fiction

With the “AI in the Loop” project, DLR, Mittweida University of Applied Sciences and IAV are working together to find solutions for the agile development of AI-based software products in the future. In this interview, Prof. Dr. Frank Köster from the DLR Institute for AI Safety and Security talks about why the project is so important.

March 2022, Prostep Newsletter

Automated driving is not the answer in every situation

Prof Frank Köster is the founding director of the Institute of AI Safety and Security at the German Aerospace Centre (DLR) and one of the coordinators of the joint project SET Level, which is concerned with the safety of automated driving. In this interview, he explains the importance of simulation and artificial intelligence for the development of automated driving vehicles.