This is a weekly newsletter that lists newly announced training programs addressing existential risk from AI. Visit AISafety.com/training for the full directory of upcoming programs.
Immersive workshop
Gen Stream: Cohort 2
Oxford, UK
7 days · Starts 22 Nov 2026
For generalists (operators, builders, communicators, field-builders) moving into AI safety. Participants build a project sourced from an AI safety org, with talks, red teaming, and mentor 1-1s – ending with a demo day.
Expenses covered
Full-time
Entry bar: high
Focus: general
Apply by 11 Nov 2026
Fellowship
PRISM AI Safety Research Fellowship: 2027
Online
4 months · Starts 31 Jan 2027
Research program by Women Who Do Data (W2D2) where expert mentors guide teams of 4–5 from research question to paper submission. Projects span technical and governance topics – ending with a final poster showcase.
No stipend
Part-time
Entry bar: mid
Focus: technical, governance
Applications not yet open
Fellowship
Anthropic Residency
Online & San Francisco, USA & New York City, USA & London, UK
Starts flexibly, after offers in January 2027
For professors in quantitative disciplines, technical founders, and staff-level engineers moving into AI research. Includes a mentored project, plus self-study time and talks to learn modern ML and AI safety.
Stipend included
Full-time
Entry bar: high
Focus: technical
Apply by 22 Nov 2026
Course
CLR Fundamentals Program: Autumn 2026
Online
1 month · Starts 20 Nov 2026
Course by the Center on Long-Term Risk (CLR) on how transformative AI might be involved in creating large amounts of suffering (s-risks). Covers model personas, safe Pareto improvements, and multiagent AI safety.
No stipend
Part-time
Entry bar: low
Focus: technical, governance
Apply by 16 Nov 2026
Fellowship
Pathfinder Fellowship: Spring 2027
Online
5 months · Starts 6 Dec 2026
Program by Kairos for students who lead or plan to found university AI safety or AI policy groups. Fellows are paired with experienced organizers and get group funding, workshops, and a Slack community.
Stipend included
Part-time
Entry bar: low
Focus: general
Applications not yet open
Course
AGI Strategy
Montréal, Canada
1 month · Starts 23 Nov 2026
Run by AI Safety Montréal (AISM) for anyone who wants to upskill in AI safety and get a clearer view of the field and their own place in it. Uses BlueDot Impact's AGI Strategy curriculum. Sessions in English or French.
No stipend
Part-time
Entry bar: low
Focus: general
Apply by 16 Nov 2026
Course
AI Safety Technical Fundamentals
Montréal, Canada
3 months · Starts 21 Nov 2026
Guided study of the ARENA materials, run by AI Safety Montréal (AISM). Covers deep learning fundamentals, transformers and mechanistic interpretability, reinforcement learning, and model evaluations.
No stipend
Part-time
Entry bar: low
Focus: technical
Course
Frontier AI Governance course
Montréal, Canada
5 weeks · Starts 17 Nov 2026
Program by AI Safety Montréal (AISM) for anyone who wants to work on how frontier AI is governed. Uses BlueDot Impact's curriculum to assess the evidence, map institutions and power, and plan how to contribute.
No stipend
Part-time
Entry bar: low
Focus: governance
Apply by 10 Nov 2026
Bootcamp
LENS Bootcamp: Learning to Explain Neural Systems
New Delhi, India
8 weeks · Starts 21 Nov 2026
Program by Secure AI Futures Lab (SAFL) on looking inside small open language models. Covers the logit lens, steering vectors, activation patching, and probes, with teams of three testing whether a model uses a concept.