AI Safety Fundamentals
Structured courses on AI alignment and governance, with curated reading lists you can also follow on your own.
Visit
Resources
Resources
A curated collection of useful links to learn about AI safety, find courses, communities, and opportunities.
Start here
Lots of Links
A large, regularly updated directory of AI safety resources: courses, communities, funding, jobs, and reading lists all in one place.
๐ Start learning
๐ฌ Research & reading
๐ Fellowships & research programmes
MATS
ML Alignment & Theory Scholars: a 12-week research programme in Berkeley and London pairing scholars with mentors in alignment, governance, and security.
Visit
Pivotal Research Fellowship
A paid 9-week in-person fellowship in London pairing fellows with mentors on AI safety, AI governance, and biosecurity research.
Visit
SPAR
The Supervised Program for Alignment Research: a part-time, remote, three-month programme pairing aspiring researchers with established mentors.
Visit
MARS
Run by the Cambridge AI Safety Hub, a part-time programme where teams of two to four work with a mentor to produce published AI safety research.
Visit
ERA Fellowship
A fully funded 10-week summer fellowship in Cambridge for researchers working on mitigating risks from frontier AI.
Visit
LASR Labs
A 13-week full-time technical research programme in London where small teams write and publish an academic paper under supervision.
Visit
Astra Fellowship
Constellation's fully funded five-month fellowship in Berkeley, with empirical research and strategy and governance tracks.
Visit
PIBBSS Fellowship
Run by Principles of Intelligence, a three-month interdisciplinary fellowship for researchers from fields that study complex and intelligent systems.
Visit
GovAI fellowships
The Centre for the Governance of AI runs three-month seasonal fellowships in London and Washington DC, plus longer research scholar and fellow positions.
Visit
Talos Fellowship
A European AI policy fellowship placing participants in EU institutions in Brussels, including people without prior tech policy experience.
Visitโก Bootcamps & intensives
ARENA
An in-person programming bootcamp in London for aspiring AI safety researchers with strong Python and maths foundations.
VisitML4Good
Eight days of free, intensive, in-person AI safety training, run as regional cohorts across Europe and beyond.
Visit
AI Safety Camp
A virtual research incubator that helps participants sharpen a confusion about AI risk into a rigorous research direction.
Visit๐ Careers & opportunities
80,000 Hours: AI risk
Career guidance and a detailed problem profile on reducing risks from advanced AI.
VisitOur partners and organizations like ours
We do not work alone. In Italy we collaborate with partner organizations, and beyond our borders AI safety groups are forming country by country: these are the people doing work like ours elsewhere.
๐ฎ๐น Partners in Italy
CePTE
Italy
Centro per le Politiche sulle Tecnologie Emergenti, an independent think tank producing evidence-based reports and policy recommendations on emerging technologies for Italian decision-makers.
Visit
Effective Altruism Italia
Italy
Altruismo Efficace Italia APS grows the Italian effective altruism community with educational programmes, career guidance, and events on how to do the most good, existential risks included.
Visit๐ช๐บ Europe
ENAIS
Europe
The European Network for AI Safety, connecting researchers and policymakers for safe AI across the continent.
Visit
Safe AI Germany (SAIGE)
Germany
A national research and field-building initiative strengthening Germany's contribution to reducing risks from advanced AI.
Visit
Dutch Network for AI Safety
Netherlands
Connects researchers, practitioners, and institutions to grow a grounded AI safety community in the Netherlands.
Visit
Safe AI Netherlands (SAIN)
Netherlands
A Dutch non-profit uniting researchers, students, and professionals around the safe development and deployment of AI.
Visit
CeSIA
France
The French Center for AI Safety, working on technical research, university-level teaching, and policy to build a culture of AI safety in France and Europe.
Visit
AI Safety Paris
France
A community bringing together people in Paris working on AI safety, alignment, and governance.
Visit
AI Safety Espaรฑa
Spain
A hub for AI safety in the Spanish-speaking world, helping people direct their careers towards building safer AI.
Visit
AI Safety Hungary
Hungary
Empowers students and researchers in Hungary through seminars, fellowships, and career support.
Visit
AI Safety Polska
Poland
A community of scientists, activists, and enthusiasts reducing risks from advanced AI through talks, meetups, and talent development.
Visit๐ Around the world
AI Safety ๆฑไบฌ
Japan
A Tokyo-based community running talks and study groups on AI safety.
Visit
AI Alignment Network (ALIGN)
Japan
A Japanese non-profit building an ecosystem for AI alignment researchers and practitioners across academia and industry.
VisitAI Safety Brazil
Brazil
A non-profit connecting professionals, researchers, and policymakers to advance safe and responsible AI development in Brazil.
Visit
BAISH
Argentina
The Buenos Aires AI Safety Hub, offering technical courses, governance programmes, and a route into AI safety research.
Visit
AI Safety Mexico
Mexico
Research, training, and community building on AI safety from a Mexican and Global South perspective.
VisitAI Safety India
India
Training, research, and a cohort of practitioners advancing AI safety in India and connecting them to the global field.
Visit
AI Safety UAE
United Arab Emirates
Delivers AI safety training, model audits, and governance work while building a community around responsible AI adoption.
Visit
AI Safety South Africa
South Africa
A Cape Town hub running community events, capacity building programmes, and a research group.
Visit
AI Safety Morocco
Morocco
Morocco's first in-person AI safety community, based in Casablanca, running meetups, reading groups, and research tracks.
Visit
AI Safety Zambia
Zambia
Works on red teaming frontier models, monitoring AI harms, and building governance and research capacity across Africa.
Visit
AI Safety Nepal
Nepal
A national AI safety community building local interest and capacity in Nepal.
Visit
Open Community for AI Safety China
China
The largest Chinese-language AI safety community, covering AI risks, safety techniques, and governance.
VisitKnow a resource we should add?
Get in touch and help us grow this list for the Italian AI safety community.