InnoLOG: 8th Thuringian AI Forum October 6, 2026 | COMCENTER Brühl | Erfurt

Take this opportunity to discuss exciting issues and share your own experiences. Cross-industry dialogue raises questions, provides answers, motivates, stimulates, and connects people.

In the moderated group discussions of the cross-industry dialogue, developers, decision-makers, users, and interested parties from various fields come together to discuss current issues surrounding the use of AI and the associated challenges. How can Thuringia become an AI hub? How can AI be integrated into existing processes so that it helps rather than hinders progress? What risks does the use of AI entail, and how can they be addressed? How will AI support the mobility of tomorrow?

The use of AI models—especially language models from major vendors—is not feasible in every application context. However, operating large models on in-house or trusted infrastructures requires significant resources and may not meet all requirements. Small yet powerful AI models for use on hardware with limited resources, such as edge devices, appear promising. But which models are the right ones for a given application context, and how can they be used effectively? This workshop provides insights into typical methods for compressing large AI (language) models. Together, we will explore how we can utilize these models, whether they are suitable for our local conditions, and how we can test models for our specific use cases.

LLM-based chatbots have become the de facto standard for text-based first-level support on websites in recent years. These chatbots typically rely on a Large Language Model (LLM) that accesses one or more knowledge sources in the form of vector databases or knowledge/context graphs. However, since these knowledge databases change over time and thus influence chatbot output, continuous monitoring of this output has become increasingly important. In this workshop, Prof. Dr. Oliver Mothes will present the open-source framework RAGAS, which offers various metrics for evaluating chatbots. These metrics can be used not only for quality monitoring but also for optimizing chatbot hyperparameters.

Most business owners have ChatGPT open. But for their own business, it rarely amounts to more than a pleasant conversation. The reason is almost always the same: the transition from idea to application is missing. Where do I start? With what? And is it even worthwhile for a business like mine? We demonstrate precisely this transition live in the workshop. At Batix, we know this scenario from years of project work. An Excel spreadsheet, a process that eats up time every month, and the question of how to actually turn it into a software project. We bring this experience to the workshop. We know how a prototype can be created in a very short time because we do it constantly in real client projects. In the workshop, we reverse this approach. Participants bring their own problem – a spreadsheet, a repetitive task, a frustrating process. Together, we select one or two examples and build a first working prototype right there on the spot. At the end of the 90 minutes, we have a tool that actually works and can be tested. Participants experience something that no lecture can convey. It's about how quickly a real, everyday problem can become a real, working solution. And how little is needed at the beginning.