Who We Are

We believe AI should be built to genuinely want to do what's best for us.

It turns out the research that gets us there also makes AI better.

We find and support the people making that happen.

The Challenge

AI systems are already showing deceptive behavior, resisting shutdown, and rewriting their own code to stay alive. The guardrails we put on these systems are surprisingly easy to break. They teach the AI what to say without changing how it thinks. Yet, we're already deploying these systems in military and critical infrastructure.

Soon, AI systems will be able to improve themselves, learning and changing on their own. If today's guardrails barely hold on systems that stay the same, what happens when the systems start rewriting themselves?

We need AI that is trustworthy by design. Systems that become more reliable, and more good, as they get smarter.

Our Approach

We find brilliant researchers working on neglected, high-potential approaches, and give them everything they need to succeed.

The biggest breakthroughs tend to come from people working ahead of consensus, on ideas that sound strange until they become obvious.

Some are outside the alignment field entirely, bringing fresh perspectives from other domains. Others are already in the space but stuck working on only what their current funding allows. We back those people. We give them engineering teams, compute, and support so they can move fast.

Team

Judd Rosenblatt

President

Zeke Reffe-Hogan

Lead Operations

Noah Maier

Lead Development

Our Partners

Alignment research moves faster when researchers have dedicated engineering support. We work with partners who bring machine learning engineering capacity into alignment research: teams, compute, and infrastructure that let researchers focus on the science.

Our founding partner is AE Studio, a technology studio and major donor whose team has been instrumental in standing up the Foundation and its research programs.