AI Act readiness · gap analysis

AI Act readiness and gap analysis

Bring AI systems, roles, risks and missing evidence into one decision-ready roadmap. In 2 weeks we combine AI inventory, classification, vendor/DPIA/FRIA signals and leadership priorities.

Readiness sprint

2 weeks

From uncertainty to route

AI register baseline and system owners

Risk classification per AI system

DPIA/FRIA, vendor and Article 4 signals

30-60-90 day roadmap for decision-making

When this gap analysis fits

This hub is for organizations that know AI Act preparation is needed, but do not yet have a clear view of systems, roles, risks and evidence.

1

No full picture yet

AI tools, SaaS features and team-level solutions are spread across IT, procurement, privacy and the business.

2

Unclear risk class

There is uncertainty around high-risk, transparency duties, GPAI, provider/deployer role or low-risk documentation.

3

Decision-making is stuck

Leadership needs to know what comes first: register, FRIA/DPIA, vendor questions, training, policy or evidence pack.

What the gap analysis delivers

AI register baseline with system, purpose, owner, supplier and process context

Risk classification per system: prohibited, high-risk, transparency duty, GPAI or low risk

Provider/deployer signal and first role split for own systems, SaaS and suppliers

Gap matrix for governance, data, logging, human oversight, transparency and monitoring

DPIA/FRIA signals for AI systems that affect people, privacy or fundamental rights

Vendor evidence and contract questions for procurement, renewal or customer conversations

Article 4 and AI literacy signals for teams that work with or manage AI

30-60-90 day roadmap with priorities, owners, decision points and logical follow-up routes

Approach in 2 weeks

1

Scope and intake

We define business units, processes, suppliers, AI systems and decision questions that fit the first readiness scope.

2

Inventory and classification

We collect systems, roles, use cases and supplier information and assign each system a first AI Act route.

3

Gap matrix

We record where evidence is missing across governance, data, vendor evidence, DPIA/FRIA, human oversight, transparency and training.

4

Roadmap

We sequence actions into 30, 60 and 90 days, with owner, urgency, dependencies and decision points.

5

Leadership session

We present findings and decide whether the next step is register, FRIA/DPIA, vendor check, Article 4 Evidence Sprint or governance framework.

Routes we explicitly include

AI register and classification

Which systems exist, who owns them and which AI Act route is likely?

FRIA/DPIA and fundamental rights

Where do systems affect people, privacy, bias, access to services or employment relationships?

Vendor evidence and contracts

Which supplier claims, role split, documentation and contract points are missing?

Article 4 and leadership reporting

Which teams need demonstrable competence and what should leadership be able to read?

Who this works for

Legal, privacy and compliance

Teams that need to organize AI Act, GDPR, DPIA/FRIA and evidence building together.

IT, security and procurement

Teams that need grip on AI tools, SaaS features, model chains, vendor evidence and contracts.

Board and leadership

Teams that need a readable priority map instead of scattered risk notes.

HR, finance, public sector and product teams

Contexts where AI influences decisions about people, services, selection, credit or customer processes.

Afterwards you know

Which AI systems need priority

Which risk and obligation route is likely per system

Which evidence is missing and who should provide it

Where DPIA, FRIA, vendor check or training is needed

Which 30, 60 and 90 day actions are realistic

Frequently asked questions

Is this legal advice or a conformity assessment?

No. This is a practical readiness and gap analysis. We structure systems, risks, signals and next steps so legal, privacy, IT and leadership can make focused decisions.

We do not know which AI systems we use yet. Is that a problem?

No. That is exactly when this is a logical first step. We use short interviews, tool overviews, procurement and privacy input to make the first register and key uncertainties visible.

How does this relate to AI inventory setup?

AI inventory setup is a compact register sprint. The readiness/gap analysis uses that register as a base, but adds classification, gap matrix, vendor/DPIA/FRIA signals and a leadership roadmap.

What if we have no high-risk AI?

That is still valuable. You then have substantiated which systems do not appear to follow a high-risk route and which duties remain, such as transparency, AI literacy, supplier control or policy.

How much internal time does this require?

Usually 2 to 4 short interviews, access to existing tool or supplier overviews and one review moment. We do most of the drafting.

Make your AI Act route concrete.

Start with the Gap Intake. That tells us which systems, suppliers and decision questions belong in the first readiness scope.

Rivium Westlaan 46, Capelle aan den IJsselCoC 90283597