Research, advisory, and investment in agentic AI safety
We are a specialized practice that researches, advises, invests, and builds initiatives that advance the foundational governance and security of agentic AI, ensuring systems are resilient to adversarial attacks, transparent, and aligned with both human judgment and organizational objectives.
Practice
Our practice is organized around the core challenges of agentic AI: governing autonomy, securing increasingly capable systems, and preserving meaningful human judgment as machines take on greater responsibility.
Contact
Tell us what you're working on.