- Ongoing
- AI Governance
A benchmark for AI agent compliance with US sanctions law
A benchmark evaluating whether AI agents comply with US sanctions law.
- Fellow:
- Arjun Sharma
- Mentor:
- Kevin Wei
- Dates:
- 18 Jan – 30 Apr 2027
- Location:
- LISA, London
- Stipend:
- £15,000–£18,000
Why do we run this fellowship?
AI might be the most consequential technology humanity builds. Whether it goes well depends in large part on how many talented people are working seriously on making it go well. The fellowship exists to grow that group.
What you get during the fellowship
Expert mentorship
Weekly 1-on-1s with established researchers providing methodological expertise, critical feedback, and strategic direction.
Research management
Pivotal's dedicated research managers help you scope projects, stay on track, navigate blockers, and plan your career.
LISA workspace
In-person co-working at the London Initiative for Safe AI. Lunch and dinner included weekdays.
Financial support
£15k–£18k stipend. Plus travel to London, £1,800/month housing for fellows based outside of London, and the compute you need for your research to succeed.
Workshops and speakers
Q&A sessions with leading domain experts and hands-on workshops to sharpen your thinking and expand your network.
15-week programme
A focused, full-time programme with sustained support to take your research from a promising idea to meaningful progress.
Where fellows go
Past fellows went on to work at
GovAI, SaferAI, UK AISI, IAPS, AI Futures Project, Anthropic Fellowship, Timaeus, Google DeepMind, Cooperative AI Foundation, Beneficial AI Foundation, FLI, the Institute for Progress, among others.
Founded
PRISM Evals, Catalyze Impact, Moirai, and Principia. More to come, stay tuned.
Started PhDs at
University of Oxford, Stanford University, EPFL, Max Planck Institute, among others.
Published research
In top academic venues and policy forums.
Our track record speaks for itself
- 8Cohorts completed
- 129+Alumni
- 9.1/10Peer recommendation by fellows
From our fellows
“Pivotal helped me grow my professional network and launch my career in AI governance.”
“The fellowship allowed me to work with top AI safety researchers — a great privilege early in my career. People were surprised what can be accomplished in two months, including me.”
“I'm more bullish on starting projects or a startup compared to research since starting at Pivotal. [I am also] more confident now that AI governance is the most important thing.”
“I thought I would like it and it would be good for my career, but I actually liked it more than I thought and think it was better for my career than I thought.”
“I think the Pivotal research fellowship accelerated my AI safety career by 15 months in expectation.”
“The fellowship sharpened my thinking on AI security and helped me figure out where I can have the most impact in AI policy.”
“Tremendously educational, inspirational, and fun. I'd highly recommend joining — it was one of the most enjoyable work environments of my life.”
“My first independent AI safety project since university. I was intimidated, but the team was supportive, and I surprised myself by how much I achieved.”
Mentors
View All Mentors
Arthur Tellis
Senior Technology Fellow, Institute for Progress
Caspar Kaiser
Associate Professor, University of Warwick
Cecilia Elena Tilli
Founder and Director, Allelo
Connor Dunlop
Head of Policy & Strategy, Lucid Computing
Daniel Tan
Research Lead, Arcadia Impact
Dave Banerjee
Researcher, Institute for AI Policy and Strategy
David Manheim
Director of Policy and Research, ALTER
Edward James Young
Member of Technical Staff, Geodesic Research
What previous fellows have worked on
All projects- Ongoing
- Technical AI Safety
A benchmark for “deal-making” with potentially scheming models
Building a benchmark that probes whether and how models engage in "deal-making" — offering a potentially-scheming model something in exchange for revealing misalignment — with a focus on qualitatively analysing the model's reasoning about credibility, cheating and being evaluated.
- Fellow:
- Mark Keavney
- Mentor:
- Dr Francis Rhys Ward
- Ongoing
- AI Governance
A disclosure framework for AI-lab insiders
Developing a framework to help AI-lab insiders judge whether and when to disclose high-impact information.
- Fellow:
- Santeri Koivula
- Technical AI Governance
A privacy-preserving side-channel monitor for AI compute
Designing a privacy-preserving side-channel monitor for AI compute — hardware-level monitoring that could help verify how chips are being used.
- Fellow:
- Sleem Abdelghafar
- Mentor:
- Gabriel Kulp
- Ongoing
- Technical AI Safety
A simpler “cross-domain solution” for protecting model weights
Designing and prototyping a simpler, smaller-attack-surface "cross-domain solution" for securely moving data — the kind of security-critical infrastructure needed to protect model weights against sophisticated state-level attackers under the SL5 standard.
- Fellow:
- Andrew Donnellan
- Mentor:
- Guy
- Ongoing
- Technical AI Safety
A taxonomy for compositional interpretability and hierarchical concepts
Working toward a taxonomy for compositional interpretability and hierarchical concepts — top-down interpretability techniques aimed at pragmatic alignment goals.
- Fellow:
- Ethan Nguyen
- Mentors:
- Logan Riggs Smith, Thomas Dooms
Application Process
You apply to work with as many mentors as you're excited about. Shortlisted candidates move on to the video interview, a work task, and a personal interview.
Written application
You, your interests, mentor preferences. · 1–2h
Video interview
Short recorded interview. · <10m
Work task
Mentor-specific research exercise. · 1–3h
Personal interview
With the team or your mentors. · 30–45m
FAQ
Who is eligible to apply?
Anyone 18 or older. We've had fellows from ML, philosophy, policy, physics, biology. Your background matters less than motivation and ability.
What's your acceptance rate? Should I apply?
If you're reading this, probably yes. Our fellowship is selective, but that alone shouldn't put you off. If your interests and background seem like a plausible fit, applying is often worth it.
Can I bring my own project?
You need to apply to at least one mentor. That said, there's often room to shape the project together with the guidance of your mentor.
What does the application involve?
A written application (~1–2 hours), then for shortlisted candidates: a short video interview, a mentor-specific work task, and a personal interview.
Who qualifies as Senior Fellow?
Senior Fellows have significant experience, such as a PhD, industry track record, or strong publications, and receive £18,000 instead of £15,000. The fellowship experience is otherwise the same.
Can I participate remotely?
It's an in-person programme in London. We make rare exceptions for people with a strong independent research track record. If you can't fully be in-person, apply either way and flag it in the form.
Can I participate part-time?
The fellowship is full-time. Small side commitments are usually fine and you can mention them in your application.
Can fellows work in teams?
Some mentors have fellows work in teams, others work one-on-one. Either way, collaboration between fellows is common and encouraged.
What outputs are expected?
Usually a research paper, but we're open to blog posts, policy briefs, or other formats if they suit the project. In any case, you retain ownership over your research.
Will I be employed by Pivotal? Do I need a UK visa?
No, Pivotal will not employ you, but we'll have a stipend agreement. Meeting entry requirements is your responsibility — we have an info doc on entry requirements and can sometimes help with the process.
What about housing and travel?
We cover travel to and from London. If you don't already live here, you'll get £1,800/month towards accommodation on top of the stipend.
Why does Pivotal run a fellowship?
The field needs more good researchers working on AI safety and governance. The fellowship is the main way we try to make that happen.
Can I work with Pivotal in other ways?
Yes, have a look at Get involved on our About page, or just get in touch.