Help 100 Professionals across the world transition into career in AI safety

We want to help at least 100 top talents across the world to transition into full-time careers in AI safety/governance and help make AI safe for all

Simulating Non-Dual Experience in LLMs via Activation Steering

Testing a geometric model of consciousness through mechanistic interpretability

The Odyssean Process for AI

Combining, iterating and refining best-practice epistemic democratic AI Governance with a documentary and international strategy to scale a UK citizen assembly

Testing evaluator self-confirmation in revisable AI systems

A reproducible multi-model study of proxy exploits, evaluator defects, distribution shifts, and withheld tests

NadIntellect Lab: Testing Constrained AI Control

Testing whether powerful AI systems can remain useful while their ability to affect the external world is structurally constrained.

Hong Kong Hub for Apart Research's AI Collusion Sprint

AISHK will run an in‑person weekend sprint hub for elite finance talent and host the winning team during the Apart Fellowship.

Mycelium: AI alignment for nonhuman welfare

Robustifying Alignment Techniques to Survive RLVR and Continual Learning

We want to measure & prevent drift in model alignment characteristics during RLVR and continual learning.

Seen & WithMe

Keep AI Whistleblowers Alive

Defining dangerous DNA with the Sequence Biosecurity Risk Consortium (SBRC)

Defining which sequences are dangerous enough to screen for, so providers and regulators screen consistently

Sentient Futures Residency

Animal Welfare Midtraining Data Creation

Toward a Superhuman, Open-Ended Alignment Auditor

AI Safety Hong Kong (AISHK)

AI Kurd - AI Safety & Alignment Education

Scaling open-source AI safety content, localized research breakdowns, and web tools for the Kurdish community specially and world generally

Auditability of deployed AI models: what can and cannot be verified from outside

A negative result on behavioural testing, a narrow positive one on cryptographic execution proof, and 8 independent attestations anyone can check

The Enduring Path to Alignment: Inorganic Existential Social Theory

Prediction of Inoculation Prompt Side Effects

An open, cheap method that detects when an inoculation prompt inoculates against off-target traits, so labs and developers can catch undesired trait/persona cha

Preserving the Human Veto

Civil-society infrastructure against AI-enabled power concentration, built around autonomous weapons

Sentient Futures Project Incubator

Funding compute/API costs for Incubator projects that build nonhuman welfare consideration into AI safety work

AI Safety x Content Fellowship

A three-week virtual fellowship where young Africans learn about AI safety and turn big ideas and catastrophe scenarios into videos the public can understand.