The First Actionable Interpretability Workshop at ICML 2025

https://actionable-interpretability.github.io/

GENESIS Protocol: Building Safe, Transparent AGI

Building a transparent, symbolic AGI that runs millions of tokens/sec on CPUs, making safe, explainable AI accessible to everyone.

Acausal research and interventions

Making sure AI systems don't mess up acausal interactions

The AI Safety Research Fund

Creating a fund exclusively focused on supporting AI Safety Research

Keep Apart Research Going: Global AI Safety Research & Talent Pipeline

Funding ends June 2025: Urgent support for proven AI safety pipeline converting technical talent from 26+ countries into published contributors

Demonstration of LLMs deceiving and getting out of a sandbox

Asterisk AI Blogging Fellowship

Implicit planning in LLMs Paper

Reproducing the Claude poetry planning results quantitatively

The First Workshop on Mechanistic Interpretability for Vision

'Making God': a Documentary on AI Risks for the Public

Geoffrey Hinton & Yoshua Bengio Interviews Secured, Funding Still Needed

Synthetic pretraining to make AIs more compassionate

Enabling Compassion in Machine Learning (CaML) to develop methods and data to shift future AI values

Scaling AI safety awareness via content creators

4M+ views on AI safety: Help us replicate and scale this success with more creators

Mox, a coworking & events space in SF

For AI safety, AI labs, EA charities & startups

Out of This Box: AI Safety Musical

Out of This Box: The Last Musical (Written by Humans)

Split Personality Training

Revealing Latent Knowledge Through Personality-Shift Tokens

Coordinal Research: Accelerating the research of safely deploying AI systems.

Funding for a new nonprofit organization focusing on accelerating and automating safety work.

Suav Tech, an AI Safety evals for-profit

General Support for an AI Safety evals for-profit

AI Governance Exchange (focus on China, AI safety), Seed Funding

Building bridges between Western and Chinese AI governance efforts to address global AI safety challenges.

Cash runway while I go through interviews/wait for OpenPhil's grant decision

I've self funded my ramp up for six months and interview/grant processes are taking longer than expected.

Concept-anchored representation engineering for alignment

New techniques to impose minimal structure on LLM internals for monitoring, intervention, and unlearning.