Chris Canal
Enabling rapid deployment of specialized engineering teams for critical AI safety evaluation projects worldwide
David Rozado
An Integrative Framework for Auditing Political Preferences and Truth-Seeking in AI Systems
Faisal Moarafur Rasul
A media and learning platform exploring how AI thinks — featuring Philosopher AI, an educative system that explains its reasoning.
Jade Master
Developing correct-by-construction world models for verification of frontier AI
Armon Lotfi
Multi-agent AI security testing that reduces evaluation costs by 10-20x without sacrificing detection quality
Jared Johnson
Runtime safety protocols that modify reasoning, without weight changes. Operational across GPT, Claude, Gemini with zero security breaches in classified use
Orpheus Lummis
Seminars on quantitative/guaranteed AI safety (formal methods, verification, mech-interp), with recordings, debates, and the guaranteedsafe.ai community hub.
Michaël Rubens Trazzi
Funding gap to pay for a video editor and scriptwriter
Ethan Nelson
Leveraging a 23K Subscriber Channel to Advance AI Safety Discourse
Justin Olive
Funding to cover our expenses for 3 months during unexpected shortfall
Chris Wendler
Help fund our student’s trip to NeurIPS to present his main conference paper on interpretable features in text-to-image diffusion models.
Quentin Feuillade--Montixi
Funding to cover the first 4 month and relocating to San Francisco
Leo Hyams
A 3-month fellowship in Cape Town, connecting a global cohort of talent to top mentors at MIT, Oxford, CMU, and Google DeepMind
Thane Ruthenis
Research agenda aimed at developing methods for constructing powerful, easily interpretable world-models.
Aditya Arpitha Prasad
Practicing Embodied Protocols that work with Live Interfaces
Sean Peters
I'd like to explore a research agenda at the intersection of time horizon model evaluation and control protocols.
Micheal smith
A National AI Co-Pilot for Emergency Response
20 Weeks Salary to reach a neglected audience of 10M viewers
Selma Mazioud
Attending NeurIPS and the San Diego AI Alignment Workshop to advance research on neural network safety and complexity.
Aditya Raj
Current LLM safety methods—treat harmful knowledge as removable chunks. This is controlling a model and it does not work.