Integrating AI into chemical safety assessment – Opportunities, challenges, and the path forward

16-17 October 2024

Sophia Antipolis, France

While AI opens a world of possibilities for chemical safety assessment, the rapid integration of its technologies presents significant challenges including concerns over the quality of training data, the dynamic nature of learning algorithms, and the implications for regulatory assessments.  

This workshop looked at:

• what is the current “state of the art” as well as Success Stories and Lessons Learned, including opportunities and risks   
• how AI could be applied in mainstream chemical safety assessment, both with respect to hazard identification and safety assessment, while minimizing risks   
• challenges associated with application of AI in chemical safety assessment

Download the programme

	Presentation (click on the title to view the slides)	Speaker	Video
1	Demystifying AI: A brief introduction	Hua Qian (ExxonMobil Biomedical Sciences, US)	Watch
2	From Data to Decisions: AI's Role in Modern Toxicology	Nicole Kleinstreuer (National Institutes of Health (NIH), US)	Watch
3	Introducing AI in EFSA systematic reviews	Fulvio Barizzone (EFSA)	Watch
4	FAIR data and data that is Fully AI Ready	Erik Schultes (GO FAIR Foundation, NL)	Watch
5	AI-assisted Next Generation Risk Assessment and Safe and Sustainable by Design Workflows enabled by FAIR Data and Knowledge	Barry Hardy (Edelweiss Connect GmbH, CH)	Watch
6	Application of AI to cell imaging for drug discovery and diagnosis	Paul Rees (Swansea University, UK)
7	AI in toxicology – The now, the new and the next	Thomas Hartung (John Hopkins University, US)	Watch
8	Generative AI for Toxicology and Drug Safety	Weida Tong (US FDA)	Watch
9	Leveraging AI for Toxicity Prediction and De Novo Compound Design Using Cell Painting Data	David Rouquié (Bayer CropScience, FR)	Watch
10	Leveraging the use of AI in the Virtual Human Platform for Safety Assessment (VHP4Safety) project	Anne Kienhuis (RIVM, NL)	Watch
11	Applications of machine learning and AI approaches to develop PBPK and QSAR models to predict ADMET properties to aid chemical safety assessment	Zhoumeng Lin (University of Florida, US)	Watch
12	An application example – The Axiom Tool	Alex Beatson (Axiom)	Watch