AI Governance

Enterprise buyers are requiring AI governance documentation from vendors. Boards are asking what oversight is in place. The EU AI Act, dozens of US state AI bills, and FTC enforcement are setting deadlines. We design frameworks, conduct audits, and build the documentation to answer all of it.

We earned Certified AI Auditor credentials in 2022, months before ChatGPT launched, and built the world's first Fairly Trained LLM. Our governance work spans ISO 42001, NIST AI RMF, and EU AI Act compliance for enterprises, AI-native companies, and PE/VC portfolios.

Starting at $25K | 1-8 weeks

EU AI Act enforcement timeline

The EU AI Act is not a single deadline. It is a rolling series of obligations that began in February 2025 and continues through 2027. Organizations that wait for the high-risk deadline will find themselves scrambling.

EFFECTIVE AI Literacy Feb 2025 EFFECTIVE Prohibited Practices Aug 2025 EFFECTIVE GPAI Obligations Aug 2025 KEY DEADLINE High-Risk AI Systems Aug 2, 2026 FULL SCOPE All AI Systems Aug 2027 TODAY Already in effect Upcoming deadline Future enforcement

Services

AI Governance Framework Development

Policies, procedures, and oversight structures for responsible AI use. Aligned to ISO 42001, NIST AI RMF, and EU AI Act requirements.

4-8 weeks

Independent AI Audit

Third-party evaluation of AI systems for bias, risk, compliance, and performance. First-cohort Certified AI Auditor credentials. No platform to sell, no conflicts of interest.

2-4 weeks

EU AI Act Readiness Assessment

Risk classification, gap analysis, and conformity assessment preparation for high-risk AI systems. Covers documentation, quality management, and post-market monitoring requirements.

2-4 weeks

AI Risk Assessment

Systematic evaluation of AI risks across your organization: model risk, data risk, vendor AI risk, and regulatory exposure.

1-2 weeks

Board & Executive AI Education

Workshops for boards and leadership teams on AI oversight obligations, risk management, and strategic opportunity. Practical, not theoretical.

Half-day to 2-day sessions

Why us

Deep technical AI expertise

LLM architecture, training data provenance, embedding models, prompt injection risks, model drift detection, agentic system design. We published the research on how to build and govern AI agents. Most governance firms can tell you what ISO 42001 says. We can tell you whether your model's evaluation methodology actually measures what it claims to.

Research-backed, not checkbox-driven

Published in Science, Phil. Trans. Royal Society A, and ACL with over 4,000 academic citations. Peer-reviewed papers on LLM evaluation, agent design, and AI governance, not recycled framework summaries. Our recommendations come from building and studying these systems firsthand.

One team, no handoffs

CPA + CIPP/US + CIPP/E + Certified AI Auditor, and the same people who understand LLM architecture and training pipelines. There is no "governance team" passing findings to a "technical team" for interpretation. The people who assess your AI systems understand both the regulation and the engineering, so nothing gets lost in translation.

Why licens.io?

AI depth

Big 4

Pivoting to AI

licens.io

Built LLMs from scratch (KL3M)

Credentials

Big 4

Generalist consultants

licens.io

CPA + CIPP/US + CIPP/E + Certified AI Auditor

Independence

Big 4

Often sell AI platforms

licens.io

No vendor conflicts, audit only

Speed

Big 4

6-12 weeks

licens.io

2-4 weeks

Pricing

Big 4

Hourly, $150K-500K

licens.io

Fixed-fee, $25K-75K

Research

Big 4

Marketing whitepapers

licens.io

Published in Science, 4,000+ citations

Who this is for

  • Enterprises deploying AI that need governance frameworks before the EU AI Act high-risk deadline
  • AI-native companies seeking ISO 42001 readiness or Fairly Trained certification
  • PE/VC portfolio companies requiring AI governance assessments across their holdings
  • Boards and leadership teams that need practical AI literacy and oversight guidance
  • Organizations using third-party AI that need vendor AI risk assessments

Frequently asked questions

What is the EU AI Act and does it apply to US companies?

The EU AI Act regulates AI systems placed on the EU market or whose outputs are used in the EU. If your AI system affects EU users or customers, it likely applies regardless of where your company is headquartered. High-risk system requirements enforce August 2, 2026.

What is ISO 42001 and how is it different from ISO 27001?

ISO 42001 is the international standard for AI management systems. It governs how organizations develop, deploy, and manage AI. ISO 27001 covers information security. Many organizations will need both, and we help with both.

Who qualifies as an independent AI auditor?

An independent AI auditor has no financial interest in the AI system being audited and no platform or product to sell alongside the audit. Our team holds first-cohort ForHumanity Certified AI Auditor credentials and maintains full independence from AI vendors.

How much does an AI governance assessment cost?

Typical engagements range from $25K-$75K depending on scope and complexity, delivered in 2-4 weeks. All fixed-fee, quoted upfront.

What is the Fairly Trained certification?

Fairly Trained is an independent third-party certification confirming that an AI model's training data was obtained under proper contractual or licensing agreements, or is public domain/open license. We oversaw the process for the first LLM to receive this certification.

Get a governance assessment

We'll evaluate your current AI governance posture, identify gaps across regulatory, customer, and board requirements, and give you a prioritized remediation plan. Fixed price, defined timeline.