Home / Services / Security & Compliance / AI Governance & ISO 42001

AI GOVERNANCE & ISO 42001

AI governance that lets you ship AI fast, and prove it is safe.

AI risk assessments, policies, model inventories and ISO/IEC 42001 readiness, aligned with the NIST AI Risk Management Framework and the EU AI Act.

Free 30-minute call · Written proposal in 48 hours · NDA before we talk

AI GOVERNANCE IN SHORT

What is ISO 42001 and AI governance?

AI governance is how an organization decides which AI it builds or buys, controls its risks and proves it behaves as intended. ISO/IEC 42001:2023 is the first certifiable international standard for an AI management system. It sits alongside the NIST AI Risk Management Framework and helps you meet laws such as the EU AI Act.

WHO IT IS FOR

AI product companies

SaaS adding AI features

Enterprises rolling out copilots

Companies selling AI into the EU

Regulated industries using AI

Teams answering AI security questionnaires

Key facts

ISO/IEC 42001

Published December 2023; certifiable by accredited bodies, like ISO 27001.

NIST AI RMF 1.0

Released January 2023, built around four functions: Govern, Map, Measure, Manage.

EU AI ACT

In force since 1 August 2024, with obligations phasing in from 2025 onward.

RISK TIERS (EU)

Prohibited, high-risk, limited risk (transparency) and minimal risk.

WHY NOW

63% of breached organizations had no AI governance policy, or were still building one (IBM, 2025).

FRAMEWORKS COMPARED

ISO 42001 vs NIST AI RMF vs EU AI Act

One is a certifiable standard, one is voluntary guidance, one is law. Most companies use all three together.

ISO/IEC 42001

NIST AI RMF

EU AI Act

What it is

International management system standard

Voluntary US framework

EU regulation

Certifiable?

Yes, by accredited bodies

No

Conformity assessment for high-risk AI

Focus

Policies, roles, risk and continuous improvement for AI

Practical risk activities across the AI lifecycle

Legal obligations based on AI risk level

Who uses it

Organizations wanting proof for customers

US companies and agencies building trust

Anyone placing AI on the EU market

Best used for

Certification and customer assurance

Day-to-day risk practice

Legal compliance in the EU

OUR ADVICE

Build your AI management system on ISO 42001, run day-to-day risk work with the NIST AI RMF, and map both to the EU AI Act. If you already hold ISO 27001, much of the structure is reusable.

EU AI ACT RISK LEVELS

Where does your AI fit under the EU AI Act?

Obligations depend on the use case, not the model. The same model can be minimal risk in one product and high-risk in another.

PROHIBITED

Unacceptable risk

Practices such as social scoring and certain manipulative or exploitative AI are banned.

Status: banned since February 2025

HIGH-RISK

High risk

AI used in areas like hiring, credit, education, critical infrastructure and some medical uses.

Requires: risk management, data governance, logging, human oversight

TRANSPARENCY

Limited risk

Chatbots, deepfakes and AI-generated content that people must be told about.

Requires: clear disclosure to users

MINIMAL

Minimal risk

Most AI, such as spam filters, recommendations and internal productivity tools.

Requires: no specific obligations; good practice still applies

DELIVERY TIMELINE

How we build your AI governance

Most organizations reach an operating AI governance program in 8 to 14 weeks, and ISO 42001 certification readiness in 4 to 8 months.

01

Inventory

Find every AI system you build, buy or embed, including shadow AI. Typical: 1–2 weeks.

02

Classify

Risk-rate each use case against the EU AI Act and your own appetite. Typical: 1–2 weeks.

03

Policies

AI policy, acceptable use, roles and approval workflow. Typical: 2–3 weeks.

04

Assess

Impact assessments, data and model risk reviews for priority systems. Typical: 2–4 weeks.

05

Controls

Access, logging, evaluation and human oversight put in place. Typical: 2–4 weeks.

06

Certify

Internal audit, management review and ISO 42001 certification support.

DELIVERY TIMELINE

Typical total: 8–14 weeks to an operating program

01

Inventory

1–2 weeks

02

Classify

1–2 weeks

03

Policies

2–3 weeks

04

Assess

2–4 weeks

05

Controls

2–4 weeks

06

Certify

2–4 months

THROUGHOUT THE PROJECT

You always know where you stand

Weekly status updates

Progress, owners and dates

One accountable lead

A single point of contact, start to finish

Product-friendly

Governance that fits your release cycle

WHAT'S INCLUDED

Everything you need to govern AI with confidence

Practical governance your product and data teams can actually follow.

✓

AI system inventory

Every model, vendor and AI feature, with owners and purpose.

✓

Risk classification

Each use case mapped to EU AI Act tiers and your risk appetite.

✓

AI policy set

AI policy, acceptable use and an approval workflow for new AI.

✓

AI impact assessments

For high-impact systems, covering people, data, bias and safety.

✓

Model and vendor reviews

Due diligence for third-party models and AI tools.

✓

Evaluation & monitoring

Accuracy, bias and safety checks, with logging and alerts.

✓

Human oversight design

Clear points where people review or override AI decisions.

✓

ISO 42001 readiness

Gap analysis, documentation and support through certification audits.

You own every policy, assessment and model record. No lock-in, ever.

EXPERT INSIGHTS

What most AI governance guides won't tell you

How to govern AI without slowing your teams down.

01

Start with an inventory, not a policy.

You cannot govern AI you do not know about. Most companies find far more AI in use than they expected, including in SaaS tools.

02

Risk comes from the use case.

The same model is low risk in a writing tool and high risk in hiring decisions. Classify uses, not models.

03

Shadow AI is a data problem.

Staff pasting data into unapproved tools is the most common AI risk. A good acceptable-use policy plus approved tools fixes most of it.

04

Evaluation is a control.

Regular accuracy, bias and jailbreak testing is both a safeguard and your evidence for auditors and customers.

05

ISO 27001 gets you halfway to 42001.

The management system structure, internal audit and document control carry over directly.

06

Track the EU AI Act timeline.

Obligations phase in over several years, and dates can change. Build a program that adapts.

TOOLS & FRAMEWORK MAPPING

Governance tools that fit AI teams

We plug governance into the platforms your AI and data teams already use.

AI PLATFORMS

Azure AI Foundry

AWS Bedrock

Google Vertex AI

EVALUATION

Promptfoo

Ragas

MLflow Evaluate

GUARDRAILS

Azure AI Content Safety

Amazon Bedrock Guardrails

NVIDIA NeMo Guardrails

DATA GOVERNANCE

Microsoft Purview

Collibra

Unity Catalog

COMPLIANCE PLATFORMS

Vanta

Drata

OneTrust AI Governance

ONE PROGRAM, MAPPED TO

ISO/IEC 42001

NIST AI RMF

EU AI Act

ISO 27001

GDPR

PROOF

Use cases for AI Governance & ISO 42001

Real problems, the services we combined, and the results.

All use cases →

FINTECH

AI strategy

AI governance

AI strategy, portfolio and governance for regional banks

THE CHALLENGE

Many AI pilots, few in production, and the EU AI Act approaching.

OUR APPROACH

A value-scored AI portfolio with governance designed in from the start.

WHAT SUCCESS LOOKS LIKE

A short list of funded use cases in production and one AI risk register.

1

risk register for every AI system

Read the story →

1 / 3

RELATED INDUSTRIES

Built for your industry

AI Governance & ISO 42001 adapted to the rules, systems and customers of your industry.

All industries →

BLOG & RESOURCES

Insights on security, compliance and privacy

Practical guides for founders and technology leaders. No jargon.

View all resources →

AI FOR SAAS

AI features for SaaS: what enterprise buyers now expect

Guardrails, evaluation and data-use terms that pass review.

Akash Soni, Founder & CEO · 8 min read

MEDIA & AI

Generative AI on your catalog, with consent and rights intact

A 6-step plan for labels, publishers and broadcasters.

Akash Soni, Founder & CEO · 9 min read

HEALTH & AI

HIPAA and generative AI: using LLMs with patient data safely

BAAs, minimum necessary data and the guardrails that protect PHI.

Akash Soni, Founder & CEO · 9 min read

FAQ

AI governance questions, answered

Straight answers for leaders rolling out AI. If yours isn't here, ask us directly.

Still unsure? Talk to a security expert. 30 minutes, no pitch.

Book a free consultation

What is ISO 42001?

ISO/IEC 42001 is the first international standard for an AI management system. It sets requirements for governing how an organization develops, provides or uses AI, covering risk and impact assessments, data quality, transparency, human oversight and monitoring. Organizations can be independently certified against it.

Is ISO 42001 certification mandatory?

No, ISO 42001 is voluntary. But it is quickly becoming the way enterprise buyers and regulators expect AI providers to prove responsible AI, much like ISO 27001 for security. It also gives you a structured way to meet EU AI Act and NIST AI RMF expectations.

Does the EU AI Act apply to US companies?

Yes. The EU AI Act applies to US and Canadian companies that place AI systems on the EU market or whose AI outputs are used in the EU, even without an EU office. Obligations depend on the risk level of the AI system and whether you are a provider or a deployer.

When do EU AI Act obligations apply?

The EU AI Act entered into force on 1 August 2024. Bans on prohibited AI practices and AI literacy duties applied from 2 February 2025, and general-purpose AI model rules from 2 August 2025. Most remaining obligations, including many high-risk system rules, apply from August 2026, with some EU proposals to extend certain deadlines.

What is the NIST AI Risk Management Framework?

The NIST AI Risk Management Framework (AI RMF) is a voluntary US framework for managing AI risks across four functions: Govern, Map, Measure and Manage. It is widely used by US enterprises and federal agencies, and maps well to ISO 42001, so one program can address both.

How long does ISO 42001 take?

Most organizations are ready for ISO 42001 certification in 3 to 6 months. That includes an AI inventory, risk and impact assessments, policies, controls for data and models, and an internal audit. Companies with ISO 27001 already in place can reuse their management system and move faster.

How do we stop shadow AI?

Shadow AI is staff using unapproved AI tools with company data. We combine a clear AI acceptable-use policy, an approved AI gateway with enterprise models, data loss prevention and browser controls, plus training. The goal is to give people safe AI tools, not just block them.

Can we combine ISO 42001 with ISO 27001?

Yes, and it is the most efficient approach. Both standards share the same management system structure, so you reuse governance, risk management, internal audit and management review. Many certification bodies can audit both together, reducing audit time and cost.

RELATED SERVICES

ISO 27001 Certification Consulting →

Share one management system across both

GDPR & Data Privacy Compliance →

Assess AI that uses personal data

AI Agent Development & Workflow Automation →

Build AI agents with governance built in

MLOps & LLMOps Services →

Monitor models in production

Find out where your AI risks are, in one free call.

Book a free 30-minute call. You'll leave with your AI risk hot spots and a practical first step, whether or not we work together.

Your AI risk hot spots

A practical first step

NDA before we talk

Book your free consultation

By submitting, you agree to our Privacy Policy. We never share your details.