In the first part of our new Responsible AI tools series, Jakob Mertes reports on the presentation of the new European platform AIOD.EU in Paris.
The European AI landscape is constantly evolving. At the same time, there is growing criticism of many common AI applications and platforms that are not considered responsible and have various gaps, for example in the area of data protection. In our new Responsible AI Tools series, we take a closer look at (supposedly?) responsible AI tools. Our aim is to give you and ourselves an overview of the available alternatives. You can expect presentations (not sponsored!), field reports and interviews with experts who are actively shaping the responsible use of AI.
In the first part of the series, Jakob Mertes reports on the Innovation Lab Day in Paris, where the new European platform AI on Demand was presented.
On June 11, 2025, the Innovation Lab Day on the European AI-on-demand platform took place at the University of Paris. As a Machine Learning Engineer at ARIC, I took the opportunity to take part in this event. The AI-on-Demand platform was launched by the EU project AI4Europe to bring together a wide range of AI tools, datasets and resources and make them easily accessible.
The day began at 8:45 a.m. with breakfast and an informal get-together. It became clear that the focus was not on networking, but on practical testing and development. At 9:30 on the dot, the day got underway with two inspirational presentations from representatives of the French companies Safenai and Sigma Nova. Both companies deal with the responsible use of AI (Safenai) and the development of specialized AI foundation models for scientific topics (Sigma Nova).
Following the compact but interesting presentations, the practical work with the AI on Demand platform, or more precisely with the AI Builder, a sub-area of the platform, began. There were various thematic blocks, each of which was structured in a short introduction and a subsequent work phase for testing.
The first topic block set the task of creating a RAG application with just a few clicks in the Design Studio. The Design Studio is an environment within the AI Builder that allows elements such as language models, databases and chat interfaces to be linked without programming code. Once the elements had been linked – which was a little difficult to understand at first – the finished RAG application could be deployed and tested directly from the Design Studio.
In the following topic blocks, we had the opportunity to create MemoryGraphs and MCP servers. If you know the elements and their links in Design Studio, these applications were also created, deployed and ready for testing in a short time.
Definitions:
RAG is an AI technique that combines language models with external information to generate more accurate and up-to-date answers and reduce hallucinations.
A memory graph is a visual representation of a program’s memory state that shows connections between data structures to correct errors and optimize memory usage.
An MCP Server (Model Context Protocol Server) is a server that serves as an interface for AI agents to access external data and tools and perform actions.
In my opinion, such fast and efficient no-code solutions have so far mainly been known from large US providers such as Azure or AWS. It was therefore very impressive to see that Europe is putting itself in a good position with the AI-on-demand platform so as not to lose touch with AI. Although there is still significant room for improvement and some bugs that need to be fixed in order to become truly competitive with the commercial providers from the US, this step is already being planned with the EU follow-up project (DeployAI).
Overall, the Innovation Lab Day not only provided insights into the technical implementation of the AI on Demand platform, but also showed how European initiatives can provide practical tools for AI applications. Although the platform is not yet fully stable and mature, the remaining steps towards Europe-wide scaling seem foreseeable and well prepared.
For my work in implementing AI for SMEs, I am already taking away a very good opportunity to try out and test AI applications. I also think that the platform offers enormous potential for SMEs in Europe and for the future integration of AI.
This format is offered as part of the EDIH Hamburg with the support of the European Union and the Hamburgische Investitions- und Förderbank.
About the author
Jakob Mertes is a Machine Learning Engineer at ARIC.
