The KIDA KON conference series is an essential part of the cross-institutional cooperation of the KIDA collaborative that supports the Federal Ministry of Agriculture, Food and Regional Identity (BMLEH). The conference will take place on September 29 and 30, 2026, at the German Federal Institute for Risk Assessment (BfR) in Berlin-Marienfelde.

The free to register, on-site conference is a lunch-to-lunch event starting at 12:00 pm on September 29 with the registration. The programme begins at 1:00 pm. On 30 September, the conference ends at around 12:30 pm, and additional activities such as hands-on-sessions are planned for the afternoon, ending around 4:00 pm. The programme includes keynotes, talks and a poster session.

Programme

Abstracts

The conference offers an ideal opportunity to explore diverse applications of artificial intelligence in research and practice, as well as the current research landscape, and to connect with colleagues from agriculture, food, and environmental research.

The topics of the conference are as follows:

• AI in food and agriculture
• AI impacts on environmental and food sciences
• Digital innovations in agriculture
• Visual farming
• Computer vision
• Data quality and interoperability
• IT-infrastructure for high-performance computing
• Legal and regulatory issues
• Other

Workshops:

Linked data - fundamentals, standards, and software solutions for research data (BfR):
This workshop offers researchers an overview of the Linked Data landscape and Semantic Web fundamentals, covering open standards such as RDF, SKOS, and OWL, as well as Linked Open Data and public knowledge graphs (e.g. Wikidata).
You will be introduced to key tools and see a demonstration of the RDF4 Risk Assessment Toolkit, showing how research data in Excel and CSV formats can be converted into RDF.
The aim is to provide you with practical knowledge for using open knowledge graphs and make your research data more interoperable and semantically enriched.

Integrating large language models into AI applications (BVL):
The goal of the workshop 'Integrating Large Language Models into Applications' is to convey the potential of integrating Large Language Models (LLMs) into third-party applications and to gain practical experience with the integration. To this end, participants will be provided with a database application with an attached language model. In the workshop, participants can send the application's data to the language model and explore the diverse possibilities through various tasks.

Presentations from 29.09.2026

Marek Jacob, DWD
Keynote: AI serving weather prediction

Stefan Dehm, BfR
Trustworthy AI for food safety: an architecture for verifiable, accountable decision support

Dr Ikbel Guidara, BVL
Implementation of AI-based support systems in the risk assessment of microorganisms

Vipin Singh, BHT Berlin
SoilNet app: AI-assisted expert-level annotations of soil horizons

Dr Axel Menning, BfR
Small models, large impact: decentralized AI for risk assessment in food and feed safety

Presentations from 30.09.2026

Viola Rädle, UBA
Keynote: Data alone will not protect the environment – understanding them will

Taras Günther, BfR
KIDA research data hub: a semantic-web platform for FAIR data sharing supporting the One Health paradigm and risk assessment

Ruturaj Shelake, FLI
AI-assisted data pipeline for automated mapping and standardized XML submission to EFSA

Dr Marco Selig, DBFZ
Cost and gain of AI

Venue
German Federal Institute for Risk Assessment (BfR)
Location Berlin-Marienfelde
Lecture theatre
Diedersdorfer Weg 1
12277 Berlin

Organiser
German Federal Institute for Risk Assessment (BfR)
Max-Dohrn-Straße 8–10
10589 Berlin
Germany
www.bfr.bund.de