Research

Evaluation science for high-stakes AI.

Our research asks a single question in four different ways: can this AI system be trusted where the cost of error is a wrongful outcome?

AI Evaluation

Rigorous, domain-grounded benchmarks that measure whether AI systems are reliable enough for public safety.

Program 01→

AI Governance

Institutional methods for deploying AI under due process, accountability, and procedural justice.

Program 02→

AI Assurance

Verification and evidence pipelines that make an AI decision auditable, traceable, and defensible.

Program 03→

AI Systems

Agentic systems and simulation environments that turn research into deployable public-safety workflows.

Program 04→

The Research Map

One connected graph.

Every asset the lab produces is a node linked to the ideas around it. The map lets anyone trace an outcome back through the full provenance of the research that produced it.

Program

A long-term line of inquiry.

Paper

Peer-reviewed findings and methods.

Patent

Protected, novel mechanisms.

Project

Systems built from the research.

Dataset & Code

Reproducible, open artifacts.

Policy Impact

Adoption inside institutions.