501(c)(3) Nonprofit Research Institute

PERMITFOLIO Regulation Research Institute

A research institute building frontier legal intelligence systems: verifiable legal reasoning, computable regulation, regulatory digital twins, and benchmarks for high-stakes AI governance.

Reference Architecture

From legal systems to verified intelligence

v0.1
01
Legal Intelligence
Source-grounded reasoning, citation verification, and human-governed AI workflows.
02
Computable Regulation
Versioned rule systems, temporal logic, and regulatory digital twins.
03
Policy Simulation
Adversarial benchmarks, scenario runs, and audit-ready evidence chains.
Research Frontier

Three connected workstreams.

The institute studies how legal systems become machine-actionable, verifiable, and simulation-ready without collapsing legal judgment into black-box automation.

Lab

Legal Intelligence Systems Lab

Research on legal reasoning models, retrieval-grounded judgment, adversarial evaluation, and human-governed AI workflows.

Research Series

Make Regulation Computable

A research series on legal systems as executable, versioned, testable infrastructure.

Light Reading

Global Policy Watch

Signal intelligence on regulatory systems, market infrastructure, supervision, and AI governance.

Research Outputs

Benchmarks, simulators, and policy intelligence.

The institute turns frontier research questions into public artifacts that can be inspected, challenged, cited, and extended.

Benchmark

Compliance Hallucination Benchmark

Adversarial evaluation for fabricated citations, stale rules, hidden exceptions, and unsafe legal confidence.

evaluation asset
Digital Twin

Regulatory Digital Twin Simulator

Simulation environments for testing legal rules, market incentives, agent behavior, and supervisory feedback loops.

simulation asset
Light Reading

Global Policy Watch

Short field notes on regulatory infrastructure shifts, market supervision, digital finance, and AI governance.