Readiness pilot

A focused path from AI inventory to reviewable evidence

The public exposure check is available now. The controlled inventory, assessment, policy, supplier, evidence and assurance workflows and controlled independent-specialist reviews are available to approved, expert-supported pilot customers.

Start free assessmentSee pricing

Product demonstration · 1 minute 59 seconds

Manage the AI lifecycle

Follow a hypothetical AI system through inventory, exposure, role, classification, ownership, obligations, evidence and controlled lifecycle decisions. All organisations, people and systems shown were created for demonstration, and no client data appears.

Read the video transcript

An AI system should not move from an idea into live use simply because a team feels ready. EU AI Fit turns that journey into a controlled, reviewable process. The portfolio shows every recorded system, lifecycle stage, accountable owner, exposure position and reviewed classification.

Inside each system record, responsibility becomes active only when the nominated owner accepts it. Before progress, the workflow checks foundations such as approved role and classification, a current exposure check, an obligation and evidence plan and operational risk review.

A lifecycle change records the request, evidence, reviewer, decision and rationale. Production is not the end: controls, indicators, incidents and material changes can trigger reassessment. The result is an operating trail from discovery and pilot through production, suspension or retirement.

The platform journey

One connected system. Six practical workflows.

Explore the work in the order most organisations encounter it, or go directly to the area that matters now. The scenarios below are hypothetical and contain no client data.

01

Discover and organise

Build a governed inventory, not another spreadsheet.

Record each AI use in its operating context, connect it to accountable people and suppliers, and preserve the changes that affect the organisation's position.

Controlled pilot

In this scenario

A product company discovers AI in three different places

In this scenario, a customer-facing feature uses an external model, the support team has enabled an AI assistant and HR is trialling a screening tool.

Each use receives its own system record, accepted owner, supplier relationship and review trigger, so the business can prioritise the consequential uses without losing sight of the wider estate.

What the workflow retains

  • A maintained AI-system register
  • Accepted ownership and decision history
  • Supplier, model and material-change relationships

Readiness workspace

Keep inventory, assessments, decisions, obligations, suppliers, risks, policies and evidence connected inside one controlled workspace.

Explore Readiness workspace
02

Assess and understand

Establish the position before turning it into a checklist.

Start with purpose and context, distinguish exposure from legal classification, and retain the sources, reasoning and human decisions behind the resulting plan.

Free initial screen and controlled workspace review

In this scenario

A recruitment assistant influences which applications receive attention

In this scenario, the supplier describes the tool as administrative support, but recruiters use its ranking to decide which candidates they review first.

The exposure assessment identifies people-decision and fairness signals; the separate role and classification review examines the actual workflow, and unresolved interpretation is routed for qualified review.

What the workflow retains

  • Retained exposure answers and decisions
  • Approved role and classification rationale
  • A source-linked obligation and review plan

Exposure check

Use a public screen for the first signals, then deepen rights, data, fairness, reliability, security, oversight, transparency, supplier and lifecycle exposure in the workspace.

Explore Exposure check

Role and classification

Separately examine operator role, prohibited practices, high-risk routes, exceptions and transparency triggers with approval gates.

Explore Role and classification

Obligation tracker

Turn the approved position into applicable work with owners, dates, guiding evidence expectations and review status.

Explore Obligation tracker

FRIA readiness workflow

Screen Article 27 applicability and gather affected-person, harm, oversight, complaint, DPIA and notification facts for review.

Explore FRIA readiness workflow

Article 50 transparency

Identify the relevant trigger, prepare channel-specific disclosure controls and retain implementation and accessibility evidence.

Explore Article 50 transparency
03

Prioritise and decide

Bring the work and the reason for it into one view.

Make ownership, deadlines and decision points visible without compressing a complex governance position into an unexplained compliance score.

Controlled pilot

In this scenario

A launch decision competes with overdue evidence and supplier work

In this scenario, a product team wants to release an AI feature while a supplier condition remains open and the human-oversight evidence has reached its review date.

The Action Centre explains why the work is prioritised, shows accepted ownership and brings the unresolved condition to the appropriate decision maker before launch.

What the workflow retains

  • An explainably prioritised action queue
  • Owned target dates and acceptance history
  • Management trends and a fingerprinted report

Governance Action Centre

Bring requirements, risks, supplier remediation, specialist findings, lifecycle decisions and reassessment signals into one prioritised queue.

Explore Governance Action Centre

Controlled policy documents

Generate six policies from workspace facts, resolve marked gaps, control approvals and revisions, and export consistent PDF or Word copies.

Explore Controlled policy documents

System readiness report

Produce a versioned system view covering role, classification, obligations, evidence, operational risk, decisions and explicit limitations.

Explore System readiness report
04

Evidence and assurance

Create records another person can actually review.

Connect evidence to the exact requirement and system, distinguish stored material from accepted evidence, and preserve the scope of every report or opinion.

Controlled pilot

In this scenario

An enterprise buyer asks how human oversight works

In this scenario, the supplier has a general responsible-AI policy, but the buyer needs evidence for the particular product and release under review.

The team links approved instructions, named overseers, intervention authority, briefing records and monitoring evidence to the requirement, then freezes the reviewed response for the buyer.

What the workflow retains

  • Requirement-linked evidence with adequacy decisions
  • Controlled reports and buyer responses
  • Version-scoped specialist opinions and technical working files

Evidence workflow

Link managed files or customer-controlled references to obligations and control review, retention, legal hold, deletion and restoration.

Explore Evidence workflow

Buyer-ready evidence pack

Register buyer questions, assign evidence-referenced answers, require separate review and export a locked response record.

Explore Buyer-ready evidence pack

Annex IV working file

For reviewed high-risk provider records, generate an editable technical-documentation draft with visible gaps, immutable versions and audited export.

Explore Annex IV working file

Independent specialist review

Freeze a defined system record for secure, time-limited review and retain the specialist's credentials, sources, corrections, opinion and attestation.

Explore Independent specialist review
05

Risk, suppliers and change

Keep governance current after the initial approval.

Watch the events that can make an earlier decision stale: supplier changes, control failures, incidents, new uses, model updates and operational recovery results.

Controlled pilot

In this scenario

A supplier changes the model behind a customer-support assistant

In this scenario, the product name stays the same, but response quality declines and escalations rise beyond the team's monitoring threshold.

The model relationship identifies every relying system, the indicator breach opens review work and the team records containment, supplier challenge, corrective action and the decisions that must be revisited.

What the workflow retains

  • Supplier decisions, conditions and reassessment dates
  • Owned risks, controls and monitoring indicators
  • Incident and recovery records with approved corrective actions

AI supplier assurance

Connect suppliers, products and model versions to relying systems; issue secure questionnaires; review evidence; and retain conditions and follow-up.

Explore AI supplier assurance

Operational risk monitoring

Maintain system-specific risks, controls, residual-risk decisions, indicators, events and systematic reassessment history.

Explore Operational risk monitoring

Incident readiness exercises

Rehearse monitoring, escalation authority, notifications and decision points, then retain simulation evidence and corrective actions.

Explore Incident readiness exercises

Recovery assurance register

Plan recovery exercises, record tested scope and achieved RPO/RTO, approve the result and export a fingerprinted report.

Explore Recovery assurance register
06

People and capability

Give each person the knowledge their decisions require.

Move beyond a universal awareness course by connecting learning to responsibilities, current systems, approved operating boundaries and renewal triggers.

Controlled pilot

In this scenario

One AI system creates different responsibilities across the business

In this scenario, operators use an AI recommendation, product staff configure it, procurement manages the supplier and leadership approves the operating boundary.

Everyone receives a shared foundation, then role-relevant learning and an approved system briefing. Assessments, acknowledgements, remediation and renewals remain available for management assurance.

What the workflow retains

  • Role profiles and approved learning pathways
  • Assessment, remediation and renewal evidence
  • Versioned system briefings and management assurance

AI literacy assurance

Profile responsibilities, assign assessed pathways, monitor progress and freeze a management report with a traceable learner evidence index.

Explore AI literacy assurance

System-specific briefings

Turn approved purpose, limits, checks and escalation routes into versioned direction that named people acknowledge and receive again after material change.

Explore System-specific briefings

Expert-supported delivery

Choose how the platform becomes operating practice.

These are service models built around the platform—not extra software features. Your organisation retains accountability; EU AI Fit provides structure, challenge and continuity.

Readiness sprint

Agree the first portfolio, participants, storage model and decisions, then use guided collaboration to establish a credible inventory, assessment and owned plan.

Explore the readiness sprint

Managed AI Compliance Desk

Use the live governance record to run a controlled quarterly review of change, evidence, suppliers, incidents, literacy, decisions and follow-through.

Explore the managed desk

Built for trust

Accuracy, transparency, and control

Deterministic methodology

The classifier is a versioned, source-linked decision tree with human approval gates and a specialist reference-review workflow, rather than a black-box result.

Expert-supported review

A controlled invitation can give a qualified independent specialist a frozen, scoped system record to review. Their opinion is retained with sources and limitations; it is not certification or legal advice.

Traceable decisions

The workflow records inputs, sources, rule version, reviewer, and status behind each decision.

Portable report

The first commercial deliverable is a buyer-ready readiness report that customers can retain and review outside EU AI Fit.

Controlled pilot data

Pilot onboarding defines approved data, transfer methods, access, retention, and deletion before sensitive evidence is accepted.

Current official sources

Public guidance is dated and linked to official EU sources, with specialist review required for customer-specific conclusions.

Start now

See where you stand today

Run the free exposure check for one AI use, understand the principal signals and request a copy of the answers you supplied.

Start free assessment