Boosting the Adverse Outcome Pathway Framework with AI

AI is reshaping science - but can it also help us understand how chemicals cause harm?

A Challenge co-led by the JRC and SCAHT invites bright and creative thinkers to explore how Artificial Intelligence can accelerate Adverse Outcome Pathway (AOP) development and strengthen the bridge between mechanistic science and regulatory decision-making.

Why this challenge?

The goal of this international initiative is simple but ambitious: to explore how today’s AI tools can help scientists develop and expand AOPs - the framework that connects biological mechanisms to adverse effects relevant for chemical safety and risk assessment.

AOPs are the maps that make mechanistic toxicology understandable and usable for regulators. Yet, building them remains a time-consuming, largely manual process. We believe AI can change that - by supporting scientists in extracting evidence, formulating hypotheses, and linking mechanisms across biological levels.

Who should join?

Are you curious about how AI could accelerate discovery in toxicology and human health protection?

Do you enjoy experimenting with new tools, prompts, and workflows to solve complex problems?

Then our Challenge might be for you.

We are looking for curious minds - researchers, students, data scientists, or toxicologists - ready to think outside the box. No rigid protocols, no predefined software: just an open field for exploration, learning, and collaboration.

How to join

If this sounds exciting, submit your candidacy.

Submit your candidacy here

Registration deadline: 14 November 2025.

If selected for the challenge, you will receive the material and instructions needed to get started shortly after registration.

Familiarity with the Adverse Outcome Pathway framework (see background reading from the OECD and SCAHT) is welcome, as no introductory training will be provided, but not mandatory. We will share useful resources to help you get up to speed.

How it works

Phase 1: Exploration and Experimentation 

Participants will test and document AI tools and prompts to generate evidence or narratives relevant for AOP development. At a midway checkpoint meeting, we will exchange insights and compare approaches.

Phase 2: Refinement and Integration

Selected participants will receive additional scientific guidance to refine their results. The most promising contributions may be featured in a hackathon-style workshop hosted by the Joint Research Centre (JRC) in Ispra, Italy - or online.

Who we are

This Challenge is a joint initiative of the European Commission’s Joint Research Centre (JRC) and the Swiss Centre for Applied Human Toxicology (SCAHT), supported by partners across the global AOP and New Approach Methodologies (NAMs) community.

What’s in it for you?

• A joint publication: The most successful contributors will co-author a paper summarising the collective insights and best practices emerging from the challenge - a unique chance to make your ideas part of the     scientific record and visible to the wider community.

• Collaboration and networking: Selected participants may be invited to the final workshop for hands-on collaboration with peers and mentors at the forefront of AI-driven toxicology.

Estimated timeline

In case of questions

Contact AI4AOPgmail [dot] com (AI4AOP[at]gmail[dot]com)