Frankfurt AI Safety
Learn. Build. Impact. Together
We are committed to reducing catastrophic risks from advanced AI systems by promoting responsible AI development.
Sign up to become a MemberWhat we do
Frankfurt AI Safety is a local group dedicated to reducing the risks posed by advanced AI systems. We empower people in the Frankfurt region to engage with AI safety and develop the skills to address this pressing problem.
We do so by:
- Educating ourselves with reading groups, workshops, events, and discussions.
- Connecting with a community of people dedicated to this problem worldwide.
- Supporting each other in making long-term plans to contribute to AI Safety through our career, engagement, or advocacy.
Our First Meetup
2026 Community Projects
Join one of our hands-on projects running from April to December 2026. Which project would you like to work on?
AI Safety Explainer Series
Produce a set of short, plain-language explainers on YouTube and Substack on key AI safety concepts aimed at the general online audience within and outside of Europe.
Output by Dec 2026:
- 3 bite-sized YouTube Videos
- 5 Substack posts
AI Safety Replication Report
Reproduce important AI safety papers to validate results based on the EU AI Act guidelines and build understanding in areas such as interpretability, evaluation methods, alignment techniques.
Output by Dec 2026:
- AI Safety Replication Paper
LLM Red: State of LLM Safety 2026
Stress test existing AI models for safety failures. Focus areas include: Jail breaks, Prompt injection, Deception, Autonomous Agent failures.
Output by Dec 2026:
- State of LLM 2026 – European Evaluation Report
Become a Mentor
Are you experienced in AI safety, machine learning, policy, or a related field? Help guide the next generation of AI safety researchers and practitioners in Frankfurt.
Expression of Interest for Mentors
Follow Us
Stay up to date with our events and community