LASR Labs opens applications for 2025 AI safety summer research program in London

LASR Labs is opening applications for their Summer 2025 AI safety research program, offering researchers a unique opportunity to contribute to reducing risks from advanced AI systems. The 13-week intensive program provides participants with funding, workspace, and expert supervision while focusing on action-relevant questions addressing concrete threat models from AI. This initiative represents a significant opportunity for technically skilled researchers to work on pressing AI safety challenges in a collaborative London-based environment.

The big picture: LASR Labs is launching a full-time, in-person summer research program in London focused on AI safety research with applications closing on April 26th.

• Participants will work in teams of 3-4 under the supervision of experienced researchers to produce technical papers addressing AI safety challenges.
• The program emphasizes action-relevant research questions tackling concrete threat models for advanced AI systems.

Key details: The program runs for 13 weeks from July 28th to October 24th, 2025, at the London Initiative for Safe AI co-working space.

• Participants receive an £11,000 stipend to cover living expenses in London.
• The program also provides food, office space, and travel arrangements for accepted applicants.

Ideal candidate profile: LASR Labs is seeking applicants with machine learning engineering experience, strong research abilities, and effective communication skills.

• Competitive applicants may have conducted relevant research, have experience with LLMs, possess industry research/ML experience, or hold a PhD in Computer Science, Physics, or Mathematics.
• The program values willingness to experiment and iterate, alongside the ability to clearly articulate experimental outcomes.

Application timeline: Interested researchers must submit their applications by April 26th at 23:59 BST with final decisions coming in early June.

• Previous research from LASR Labs includes papers on feature splitting in sparse autoencoders and steganographic collusion in language models.
• For additional information, potential applicants can contact the organizers at [email protected].