AI governance

AI governance for growing companies: a clear AI policy, a list of every AI tool and agent, and controls that actually enforce it.

AI governance does not have to be a 60-page framework. We give you an AI acceptable-use policy, an AI app and agent register, an approval flow for new tools, staff guidance and the Microsoft controls that back it up, including governance for Copilot and AI agents.

Discuss AI governance
AI tools and agents passing through ownership, access and review checkpoints.
The initiative

Move from a difficult technology challenge to a system your teams can use.

For leadership teams, IT heads and compliance owners who need AI used safely across the business without slowing people down. We begin with the business context and the systems already in place, then shape a delivery path around what needs to work.

  • AI acceptable-use policy
  • AI app and agent register with owners
  • Approval flow for new AI tools
  • Copilot and AI agent governance
  • Staff AI guidance and training
  • Mapping to ISO 27001, ISO/IEC 42001 and data protection laws (readiness, not certification)
Explore relevant engineering stories
When this solution fits

Start with the work your team needs to make possible.

Most growing companies have AI everywhere and governance nowhere: no written rule, no list of tools, no owner for the AI agents people build. Customers, auditors and insurers have started asking how AI is controlled. A practical governance program answers them with a short policy, a register, an approval flow and technical controls that match what the policy says.

For leadership teams, IT heads and compliance owners who need AI used safely across the business without slowing people down. The first useful step is to define what needs to change, how the current environment shapes the work and which outcome the team needs to be able to assess.

Typical workstreams

What the engineering may involve.

The exact scope depends on your systems and constraints. These workstreams describe common parts of this solution area.

01

Policy and register

Write an AI acceptable-use policy that fits how you work, and build a register of AI tools and agents with owners, data and purpose.

02

Controls that enforce it

Back the policy with Microsoft controls: DLP for AI, app approval, Entra ID access, agent identities and audit logs.

03

Run it

An approval flow for new AI tools, staff guidance, quarterly reviews and evidence your auditors and customers can see.

Why now

Customers, auditors and insurers are asking how you control AI.

AI questions now appear in security questionnaires, vendor reviews and audit requests. A short, real governance program answers them, and protects the business while people keep using AI.

01

Customer questionnaires

Enterprise customers ask which AI tools touch their data and how that is controlled.

02

Audits

ISO 27001 and SOC 2 auditors ask how AI tools fit into access control, data protection and supplier management.

03

Data protection laws

Privacy laws such as GDPR and the DPDP Act apply to personal data whether it goes into a spreadsheet or an AI prompt.

04

AI agents

Copilot Studio and other agents act on company data. Each one needs an owner, an identity and limits.

05

Cost

AI subscriptions and usage spread across teams with no view of what is spent or why.

06

Reputation

One leaked customer file in a public AI tool is enough to lose a contract.

What the program includes

Short documents, real controls, a simple routine.

We keep it proportionate to your size. The documents are short enough that people read them, and every rule has a control behind it.

Get the free AI acceptable-use policy template

01

AI acceptable-use policy

Which AI tools are approved, what data may go into them, what needs approval and who to ask.

02

AI register

Every AI tool and agent with owner, purpose, data, users, account type and review date.

03

Approval flow

A simple request and review process for new AI tools and agents, with a risk checklist.

04

Agent governance

Agent identities and permissions with Entra ID, Copilot Studio controls and, where available, Microsoft Agent 365 (some features in preview).

05

Technical enforcement

DLP for AI, app approval in Defender for Cloud Apps, admin consent for AI add-ons and audit logs.

06

Staff guidance

A short training session and a one-page guide on using AI safely at work.

Standards

Aligned with the frameworks your customers know.

We do not certify you. We map your AI controls to the frameworks you are working towards and give your auditor clear evidence.

01

ISO/IEC 42001

The AI management system standard: policy, risk assessment, AI inventory and lifecycle controls, used as a readiness guide.

02

ISO 27001 and SOC 2

AI tools and agents brought into access control, supplier management, logging and data protection controls.

03

NIST AI Risk Management Framework

Govern, map, measure and manage, used as a simple structure for AI risk decisions.

04

Data protection laws

Personal data in AI tools handled in line with GDPR, the DPDP Act and similar rules.

Checklist

AI governance quick check

A practical starting point for a company of 20 to 500 people.

Request an assessment
A clear path from scope to operation

Keep decisions connected through delivery.

Move from a defined initiative to an operating system through focused, reviewable stages.

01

Understand the workflow

Agree on the users, business context, systems and constraints that define the initiative.

02

Shape the technical path

Connect architecture, integrations, data, security and delivery decisions to the needs already identified.

03

Build in useful increments

Deliver and review the system in stages so the work can respond to product and operational feedback.

04

Prepare to operate

Consider ownership, monitoring, support and the next improvements as part of the delivery plan.

Discuss this solution
FAQ

Questions about ai governance program

What is AI governance?

AI governance is the set of policies, roles and controls that decide which AI tools and agents a company uses, with what data, under whose ownership, and how that is checked.

Do small and mid-size companies need AI governance?

Yes, in proportion. A short policy, an AI register, an approval flow and a few technical controls cover most of the risk, and answer customer and auditor questions.

What should an AI acceptable-use policy include?

Approved tools, data that must never go into AI, rules for personal accounts, how to request new tools, human review of AI output, and who owns the policy. Our free template covers each of these.

How do you govern AI agents?

Each agent gets an owner, a purpose, a list of data sources and actions, its own identity with least-privilege access, and logging. On Microsoft 365 we use Entra ID, Copilot Studio controls and, where available, Agent 365.

Can you get us ISO 42001 certified?

We do not issue certifications. We help you put the AI policies and controls in place and prepare evidence; an accredited certification body performs the audit.

How long does it take?

A governance program usually takes three to six weeks, depending on how many AI tools and agents are already in use.