Leaders of TodayLeaders of Today
BrowseWorkspaceProfile
Leaders of Today

Every opportunity worth finding, checked against its source.

Submit an Opportunity
Leaders of TodayAbout UsBlogContactReport a Bug
LegalCookiesDisclaimerPrivacy PolicyTerms of Service

© 2026 Leaders of Today™. All Rights Reserved.

£15,000 AI Safety Research Fellowship (Winter 2027)

  • Organization: LASR Labs
  • Type: Fellowship
  • Location: Global
  • Deadline: September 20, 2026
  • Funding: Fully Funded

LASR Labs, a technical AI safety research programme supported by the UK charity Arcadia Impact, takes a cohort of researchers through one project from proposal to publication in 13 weeks. Teams of three to four work full time and in person from the London Initiative for Safe AI, each supervised by an experienced AI safety researcher. The Winter 2027 cohort runs from 11 January to 9 April 2027, and the programme covers living costs so that people can move into full-time safety research without funding it themselves. **What you get:** - A £15,000 stipend to cover living expenses in London - Food, office space and travel, provided by the programme - Visa support under the Government Authorised Exchange programme for participants who do not already have the right to work in the UK - Supervision from an experienced AI safety researcher, with past supervisors drawn from Google DeepMind, the UK AI Security Institute, Anthropic and UK universities - A desk at the London Initiative for Safe AI alongside other AI safety organisations and researchers - Talks from leading researchers, a careers fair, and visits to organisations including Google DeepMind and the UK AI Security Institute - A published output: LASR papers have been accepted at NeurIPS, ICLR and ICML workshops, including three NeurIPS 2025 main conference papers **What you'll do:** - Spend week 0 studying and critically evaluating a set of technical AI safety research projects, then state your preferences and get matched into a team of three to four - Spend the remaining 12 weeks designing and running experiments, iterating on results, and writing up an AI safety research paper with an accompanying blog post - Meet regularly with your supervisor, with LASR staff supporting research progress and planning across the programme - Work full time, about 40 hours a week, in person in London for the whole 13 weeks **Key dates:** - Applications close September 20, 2026 at 23:59 GMT - Programme runs January 11 to April 9, 2027 - Supervisor names are released a few weeks before the programme starts; projects are shared in week 0 **More information:** - The application process has three stages: an application review, then work tests, then an interview - Visa questions go to contact@arcadiaimpact.org - LASR runs three cohorts a year, so a missed deadline is not the last one

Apply for £15,000 AI Safety Research Fellowship (Winter 2027)

LASR LabsFellowship

£15,000 AI Safety Research Fellowship (Winter 2027)

Sep 20, 2026 23:59:00 GMT
Opportunity type
Fellowship
Funding
Fully Funded
Location
Global
Deadline
Sep 20, 2026 23:59:00 GMT

About this opportunity

LASR Labs, a technical AI safety research programme supported by the UK charity Arcadia Impact, takes a cohort of researchers through one project from proposal to publication in 13 weeks. Teams of three to four work full time and in person from the London Initiative for Safe AI, each supervised by an experienced AI safety researcher. The Winter 2027 cohort runs from 11 January to 9 April 2027, and the programme covers living costs so that people can move into full-time safety research without funding it themselves.

What you get:

  • A £15,000 stipend to cover living expenses in London
  • Food, office space and travel, provided by the programme
  • Visa support under the Government Authorised Exchange programme for participants who do not already have the right to work in the UK
  • Supervision from an experienced AI safety researcher, with past supervisors drawn from Google DeepMind, the UK AI Security Institute, Anthropic and UK universities
  • A desk at the London Initiative for Safe AI alongside other AI safety organisations and researchers
  • Talks from leading researchers, a careers fair, and visits to organisations including Google DeepMind and the UK AI Security Institute
  • A published output: LASR papers have been accepted at NeurIPS, ICLR and ICML workshops, including three NeurIPS 2025 main conference papers

What you'll do:

  • Spend week 0 studying and critically evaluating a set of technical AI safety research projects, then state your preferences and get matched into a team of three to four
  • Spend the remaining 12 weeks designing and running experiments, iterating on results, and writing up an AI safety research paper with an accompanying blog post
  • Meet regularly with your supervisor, with LASR staff supporting research progress and planning across the programme
  • Work full time, about 40 hours a week, in person in London for the whole 13 weeks

Key dates:

  • Applications close September 20, 2026 at 23:59 GMT
  • Programme runs January 11 to April 9, 2027
  • Supervisor names are released a few weeks before the programme starts; projects are shared in week 0

More information:

  • The application process has three stages: an application review, then work tests, then an interview
  • Visa questions go to contact@arcadiaimpact.org
  • LASR runs three cohorts a year, so a missed deadline is not the last one

Who can apply

  • No specific experience requirement is stated; applicants are assessed on technical ability, research ability and communication
  • Machine learning engineering experience and strong quantitative skills
  • Willingness to experiment, iterate and execute under uncertainty, and to build a theory of change for a project focused on impact
  • An ability to state clearly what an experiment showed and what it implies, with transparent reasoning
  • Available full time, about 40 hours a week, for the full 13 weeks; the programme is not offered part time
  • Able to be in London in person; remote applicants are not accepted, though someone who can be in person for the majority will still be considered
  • Open to applicants who are not UK nationals; those without an existing right to work in the UK are supported with a visa under the Government Authorised Exchange programme
  • Written answers must be your own; LASR asks applicants not to use LLMs on them and says it may run AI detection
  • People with machine learning engineering experience and strong quantitative skills who want to move into full-time technical AI safety research. LASR names no formal experience requirement, and past cohorts have included an undergraduate who had not yet finished his bachelor's degree, an engineer with several years of industry experience, and a chemistry PhD student moving across from another field. The organisation explicitly encourages people from underrepresented groups not to rule themselves out.

AI Verification

✓ AI Verified

Our AI read the official sources and cross-checked the key facts. Here is what held up, and what it could not confirm.

We could not fully verify this organization or all of its details. Check it independently on the official website before you apply, and never share sensitive personal or financial information until you are sure a request is genuine.

88
LOT Verification ScoreStrong

Cross-checked 18 of 21 key facts

Confirmed Our best read Not stated

Good to know

  • +£15,000 for 13 weeks, plus food, office space and travel costs.
  • +They sponsor visas, so you can apply from anywhere in the world and they will cover the visa cost to get you to London.
  • +Real research output. You work in a small team on an AI safety problem and aim at a publishable paper.
  • +Run by Arcadia Impact, a registered charity in England and Wales, funded by a grant rather than by fees.

Worth checking yourself

  • !You have to be in London in person for the full 13 weeks. They do not accept remote participants, so factor in whether you can move.
  • !They are looking for machine learning engineering experience and strong quantitative skills, so this suits someone already working in or studying the field.
  • !The programme runs 11 January to 9 April 2027, so check those dates against your studies or work before applying.

LOT finds and checks opportunities from public sources. Always confirm details on the organiser’s own page before applying.

£15,000 AI Safety Research Fellowship (Winter 2027)
Sep 20, 2026 23:59:00 GMT