We're a non-profit building Italy's AI safety community, bringing people together to learn, run events, and work on making advanced AI safe.
Mentee applications: 1 – 20 November 2026
First edition, 15 December 2026 – 15 March 2027
The program connects people who want to enter the field of AI safety, ethics and governance with people who already work in it, in Italy and abroad.
Tracks, requirements and key dates →Every other Thursday at 18:30. We start online with Enrico Bottazzi on steering a model's behaviour through mechanistic interpretability, then watermarking in AI-generated text (29 October) and geometric approaches to interpretability (12 November).
Register on LumaFirst edition from 15 December 2026 to 15 March 2027. Mentor applications close 20 October; mentee applications open 1 November.
Key datesOur first in-person meetup, at Cascina Cuccagna in Milan: an introduction to AI safety, the community's new projects, then aperitivo and networking.
Recording coming soonFirst edition, winter 2026/27: from 15 December 2026 to 15 March 2027
The AI Safety for Italy mentorship program connects people who want to enter the field of AI safety, ethics and governance with people who already work in it, in Italy and abroad. The goal is to lower the barrier to entry for those who have no one around them working on AI safety, to produce concrete work that stays with the community, and to build lasting connections between people working in this field inside and outside Italy.
Tracks, requirements and key dates → In partnership with
Centro per le Politiche sulle Tecnologie Emergenti
Students, recent graduates, PhD students, professionals and researchers, but also people with other backgrounds. No experience in AI safety is required.
AI will soon radically transform our society, for better or worse. Experts broadly expect significant progress during our lifetimes, potentially achieving human-level intelligence.
We work to ensure AI is developed to benefit humanity's future. Through education, community building, and raising awareness, we create a network of people committed to safer AI development.
About us and our approach →Reading groups, seminars and a 12-week mentorship program that connects newcomers with people already working in AI safety, in Italy and abroad.
Mentorship open call · seminarsIn-person meetups across Italian cities, an active Discord and links to the international AI safety network.
Working groups on technical safety (interpretability, evaluations, alignment) and on AI governance and policy.
Open to all members
Elisabetta Rocchetti
President
Bernardo Manfriani
Operations
Carola Caivano
Communications & partnerships
Francesco Ortu
Technical infrastructure
Lorenzo Basile
Mentorship program
Valeriia Povergo
Governance & fundraising
Edoardo Cecchinato
Seminars
Carlo Martinucci
Public outreach
Supported by
Whether you're a researcher, student, professional, or simply curious about AI safety, there's a place for you in our community.
