Mechanistic Interpretability research for unfaithful chain-of-thought (1 month)

Iván and Jett are seeking funding to research unfaithful chain-of-thought, under Arthur Conmy's mentorship, for a month before the start of MATS.

Building an AI Accountability Hub

Case Study: Defending OpenAI's Nonprofit Mission

CANCELLED 24,000 subscriber YouTube channel on AI safety | Dr Waku

CANCELLED Cover anticipated costs for making videos in 2025

Horizon Events 2025

Non-profit facilitating progress in AI safety R&D through events

AI safety fieldbuilding in Warsaw, Poland (funding for 1 semester)

Reach the university that trained close to 20% of OpenAI early employees

Finishing The SB-1047 Documentary In 6 Weeks

How California became ground zero in the global debate over who gets to shape humanity's most powerful technology

Apart Research: Research and Talent Acceleration

Support the growth of an international AI safety research and talent program

Act I: Exploring emergent behavior from multi-AI, multi-human interaction

Community exploring and predicting potential risks and opportunities arising from a future that involves many independently controlled AI systems

80,000 Hours

We provide research and support to help people move into careers that effectively tackle the world’s most pressing problems.

SEA Observatory

Keeping track of AI governance in Southeast Asia - an AI Safety Asia initiative.

Singulrr 1.0

Exploring grounds to apply quantum physics in AI

Compute for 4 MATS scholars to rapidly scale promising new method pre-ICLR

Athena 2.0

Program for Women in AI Alignment Research

Accelerating the AI Safety talent pipeline in South Africa| Matched donations

Relocating to Montreal to work full time on AI safety

Doom Debates - Podcast & debate show to help AI x-risk discourse go mainstream

Let's warn millions of people about the near-term AI extinction threat by directly & proactively explaining the issue in every context where it belongs

The AI Governance Archive (TAIGA)

Site maintenance, grant writing, and leadership handover

Next Steps in Developmental Interpretability

Addressing Immediate AI Safety Concerns through DevInterp

Developing a Course on AI x-risk

Salaries for SAE Co-occurrence Project

Working title - “Compositionality and Ambiguity: Latent Co-occurrence and Interpretable Subspaces”