Reliable, Robust, and Trustworthy AI

Trusting Artificial Intelligence

We enable the efficient development and reliable operation of trustworthy AI systems.

Symbol Human agency

Human agency

Foster users' trust and acceptance through tailored Human-AI interfaces.

Symbol Robust systems

Robust systems

Ensure reliability and resilience by building on sound foundations.

Symbol Data privacy

Data privacy

Utilize and benefit from data while keeping it protected and confidential.

Architecture diagram of a recommendation system

Case Study: An explainable LLM-based recommendation system

The capabilities of pre-trained Large Language Models (LLMs) open up new possibilities for the development of recommendation systems. However, data protection requirements, limited data availability, and last but not least the black-box nature of LLMs present hurdles for their use.

We show how Explainable AI and feedback mechanisms support self-determined use of the recommendation system. In addition, we present strategies for data protection-compliant and cost-effective operation of LLM-based solutions.

Different target groups have different needs and requirements for the explainability of AI systems

Better User Experience with Explainable AI

A structured approach for the development of explanation components has proven successful in numerous projects in research and practice.

In four consecutive phases, we first capture the target group and application context, then identify suitable XAI methods, which we then test with a prototype and finally develop into an application-ready component.

Pipeline (symbolic image)

CI/CD with GitLab: Deploying a static website to your own server

CI/CD – Continous Integration / Continous Deployment – is state-of-the-art in modern software development, and the open-source code management platform GitLab comes with an integrated CI/CD feature called GitLab CI. A good approach to kill two birds with one stone: to get to know the basic mechanics of GitLab CI and at the same time to tackle the topic of Continuous Integration and Continuous Deployment.

In the first part of the tutorial series, we work out the basics of CI/CD with GitLab using the example of a static website. It consists of some HTML, CSS, and graphic files and is provided on a web server. Even in this simple case, the advantages over manual deployment from your own computer are clearly visible.