Trust

Trustworthy artificial intelligence (AI)

AI systems are increasingly taking over decision-making, whether in industrial manufacturing processes, healthcare, or public administration. However, how can we ensure that these decisions are transparent and fair?

Fraunhofer FOKUS develops methods of explainable AI, for example, in the EU project XMANAI, to make decision-making processes transparent. 

Trust in Data 

To detect fake news, deepfakes, and manipulative content, Fraunhofer FOKUS works with linguistic AI models (Project Veritas) and on technical standards such as the C2PA protocol, which tracks the origin of digital media (FAMIUM Content Provenance & Authenticity solution).

Citizens must be able to trust that their data is protected – especially in public administration or sensitive areas such as healthcare. Only then will they use digital services.

Fraunhofer FOKUS develops solutions for trustworthy data spaces, e.g., with the PISTIS project, which enables secure data transfers. Data protection and user-friendliness are also key priorities in the development of trustworthy digital identities, e.g., with SmartID.

Transparent urban data platforms such as those in Jena or the Data Competence Center for Cities and Regions (DKSR) create trust in data-driven urban development through openness and participation.

Trust in interconnected systems

Modern IT infrastructures are highly interconnected – from the Internet of Things to cloud services to global supply chains. Trust in the reliability and security of these systems is crucial.

In the DOSS project, Fraunhofer FOKUS is working on secure IoT operations with supply chain control. In the field of autonomous and teleoperated driving (AINET project), our researchers are investigating how technological bridges – e.g., through human remote control – can create trust in new mobility concepts.

Trust also plays a central role in human-machine interaction: In a current project, we are analyzing user experience with mobile service robots – and how trust can be specifically promoted through design and behavior.

Trust is not a substitute for control

Trust in IT comes not only from intuitive use, but also from verifiable security and knowledge that enables risk assessment. This is challenging in an environment where technologies are changing rapidly. Evaluation and control mechanisms must be continuously adapted.

Fraunhofer FOKUS has extensive experience in software certification and is increasingly focusing on continuous, automated auditing to keep pace with dynamic developments. To this end, FOKUS researchers are also applying their experience in the field of cloud security (EU SEC project) to the quality assurance of AI-based software (IML4E project). In the CertLab, our experts support certification processes in accordance with international standards.