AI governance, before the questions arrive.
Customers, regulators, insurers and boards now ask how you use AI and how you control it. We build an inventory of the AI you use and build, classify it against the EU AI Act, set the policies and checks that fit your risk and, if you want a certificate, prepare you for ISO/IEC 42001.
- Founder-led
- ISACA CISM
- US & EU
- Reply within one business day
Usually, one of these is true.
This is advisory support, not legal advice. We do not issue ISO/IEC 42001 certificates; an accredited certification body does.
- Staff use AI tools and nobody has a list of which ones, or of what data goes into them.
- You build or sell AI features and EU customers are asking about the AI Act.
- A customer questionnaire, insurer or board has asked for your AI policy and risk controls.
- You want a recognised management standard for AI, alongside ISO 27001.
The AI rules at a glance
| Topic | What to know |
|---|---|
| EU AI Act | Regulation (EU) 2024/1689 applies to providers and deployers of AI used in the EU, including non-EU companies whose AI output is used there. |
| Already applying | Prohibited practices and AI literacy since 2 February 2025, rules for general-purpose AI models since 2 August 2025, and transparency duties, such as telling people they are dealing with AI, since 2 August 2026. |
| High-risk systems | After the 2026 amendments, obligations for stand-alone high-risk uses, such as hiring, credit and access to essential services, apply from 2 December 2027, and for AI built into regulated products from 2 August 2028. |
| Penalties | Up to EUR 35 million or 7% of worldwide turnover for prohibited practices, and up to EUR 15 million or 3% for most other breaches. |
| ISO/IEC 42001 | The certifiable management-system standard for AI, published in 2023. It works alongside ISO 27001 and is audited by accredited certification bodies. |
| NIST AI RMF | A voluntary US framework built on four functions, Govern, Map, Measure and Manage, with a profile for generative AI. |
From inventory to oversight that holds up.
AI inventory
Every AI system you use or build: its purpose, its data, its supplier and its owner, kept as a live register.
Classify and map duties
Each system classified against the EU AI Act, with your role as provider or deployer and the duties that follow.
Policy and controls
An AI use policy, approval steps for new tools, data rules, human oversight and vendor checks, sized to your risk.
Literacy and training
Practical training for the people who use and oversee AI, as the AI Act expects.
Certification readiness, if you want it
An AI management system mapped to ISO/IEC 42001, built on your ISO 27001 or SOC 2 work where possible.
Deliverables you can hand to an auditor.
Scoped on the free call and agreed as a fixed fee. Start with an Assess project, then Build and Manage as you need them.
- AI use register with owners and review dates
- EU AI Act classification and duties map
- AI use policy and approval process
- Vendor and data checks for AI tools
- AI literacy training plan and records
- ISO/IEC 42001 gap assessment, if you want certification
Questions about AI governance.
Does the EU AI Act apply if we only use AI tools?
Yes, as a deployer, though the duties are lighter than for providers: use systems as instructed, ensure human oversight and AI literacy, and be transparent in some cases. High-risk uses carry more.
Is our chatbot high-risk?
Usually not. Most chatbots carry a transparency duty: telling people they are talking to AI. High-risk uses are listed in the Act, such as hiring decisions, credit scoring and access to essential services.
Do we need ISO/IEC 42001?
Only if customers or your board want independent assurance. Many organisations start with the inventory, policy and controls, then decide on certification later.
What about US AI laws?
They are changing quickly and vary by state. We track what applies to you and use the NIST AI RMF as a stable base that maps to most of them.
Related: ISO 27001 certification · GDPR and US privacy
Start with a free conversation about AI governance.
A free 30-minute conversation to agree what matters most and what to do first. If you would like to prepare, answer the short questionnaire beforehand.
- Founder-led
- ISACA CISM
- US & EU
- Reply within one business day