Business Compliance Tools
Find a tool

Screen an AI use against the EU prohibited-practices routes

Record the facts behind the supported Article 5 routes and stop for specialist review when a potential prohibited-practice indicator is present.

EUAI governance
Free to useRun the complete browser-local toolAbout 15 minutes6 working filesStart this tool

Tool available. Run the complete browser-local workflow and inspect the result. Paid export is not active. Entries and files stay in this browser. Sources reviewed 7 August 2026.

Inspect a real output

Built from fictional sample information using the same exporter as the tool.

EU prohibited-practices screening sample

What’s included

Orient the reviewer

Explain the supported scope, stop conditions, future-dated item and decision limits.

  • Start Here PDF

Review and record the screening

Inspect the detailed result, editable screening register and formal decision record.

  • Prohibited-practices preflight PDF
  • Screening and action register XLSX
  • Decision record DOCX

Retain the factual basis

Keep the machine-readable screening facts and the official source and assumption record.

  • Structured screening record JSON
  • Sources and assumptions record
View every output and format
  • Start Here PDF
  • Prohibited-practices preflight PDF
  • Screening and action register XLSX
  • Decision record DOCX
  • Structured screening record JSON
  • Sources and assumptions record

Who this tool is for

Product, operations, risk, legal and AI governance teams reviewing one EU-connected AI use before deployment or material change. Create a factual first-pass screening record that does not turn uncertain scope or missing evidence into a false clearance.

Supported use

  • One user-confirmed EU-connected AI system or proposed use
  • The supported Article 5 manipulation, vulnerability, scoring, policing, facial-image, emotion and biometric routes
  • A future-dated intimate-content route that is clearly distinguished from current obligations

Have ready

  • The EU connection, user-confirmed AI-system position and organisation role
  • The deployment setting, affected people, decision or output and human involvement
  • Any exception or authorisation record being considered
  • A factual evidence reference, reviewer and review date for every screened route

Needs separate review

  • Whether the technology meets the legal definition of an AI system
  • Whether an exception, law-enforcement route or authorisation is available
  • Legal interpretation of disputed facts, territorial scope or a potential indicator

How it works

  1. Confirm the routeRecord the EU connection and the organisation position that the use is an AI system.
  2. Describe the useRecord the deployment setting, affected people, output, decision influence, human role and any claimed exception.
  3. Screen every routeAnswer each supported indicator from named factual evidence. Uncertain facts remain unresolved.
  4. Stop or recordA potential indicator produces a stop position for specialist review. Otherwise retain the dated screening record.

Where does this task apply?

Choose the guidance set this tool should use. The interface and outputs stay in English.

Using European Union guidance.

Screen an AI use against the prohibited-practices routes.

Record the use-case facts behind every supported Article 5 route. A potential indicator stops the preflight and stays assigned for specialist review.

Browser-local evidence deskSystem facts, uploaded contract text and generated records stay on this device.
1

Create the controlled record

Give this assessment a stable reference, accountable owner and review date.

2

Confirm the supported route

This tool is for a user-confirmed EU-connected AI system or use. Uncertain scope fails closed and does not produce a clear result.

3

Record the evidence for every item

Keep exact document, register, test or system references. A selected answer is not treated as complete without the evidence metadata needed for review.

Behaviour and vulnerability2 items
Harmful manipulation or deception

Could the system use subliminal, purposefully manipulative or deceptive techniques that materially impair an informed decision and cause, or be reasonably likely to cause, significant harm?

EU AI Act Article 5(1)(a)
Exploitation of a person’s vulnerability

Could the system exploit vulnerability due to age, disability or a specific social or economic situation to materially distort behaviour in a way that causes, or is reasonably likely to cause, significant harm?

EU AI Act Article 5(1)(b)
Scoring and policing2 items
Social scoring

Could the system evaluate or classify people over time based on social behaviour or known, inferred or predicted characteristics, leading to the prohibited detrimental or unfavourable treatment?

EU AI Act Article 5(1)(c)
Individual criminal-risk prediction

Could the system assess or predict the risk that a person will commit a criminal offence based solely on profiling or personality traits and characteristics, outside the supported exception for assisting a human assessment based on objective and verifiable facts?

EU AI Act Article 5(1)(d)
Biometrics and emotion4 items
Untargeted facial-image scraping

Could the system create or expand a facial-recognition database through untargeted scraping of facial images from the internet or CCTV footage?

EU AI Act Article 5(1)(e)
Emotion inference in work or education

Could the system infer emotions of a person in a workplace or education institution outside the supported medical or safety exception?

EU AI Act Article 5(1)(f)
Sensitive biometric categorisation

Could the system categorise people using biometric data to deduce or infer the sensitive attributes listed in Article 5, outside the narrow labelling or filtering exception?

EU AI Act Article 5(1)(g)
Real-time remote biometric identification

Could the system be used for real-time remote biometric identification in publicly accessible spaces for law-enforcement purposes without a separately documented permitted objective, necessity, proportionality and authorisation route?

EU AI Act Article 5(1)(h) and Article 5(2) to (7)
Synthetic intimate content1 item
Non-consensual intimate or child sexual abuse material

Could the system be placed on the market, put into service or used to generate or manipulate non-consensual sexually explicit or intimate content, or child sexual abuse material?

Regulation (EU) 2026/1744, future Article 5 prohibition
Create the screening record9 deterministic evidence items. No model output is used in the assessment.

A “no indicator identified” result records only the facts entered and reviewed. It is not a legal clearance or conformity finding.

The future intimate-content item is labelled with its 2 December 2026 application date and must not be presented as already applicable before that date.

What this tool checks

The scope stays narrow so the result is clear and reproducible.

  • Requires the user to confirm the EU connection and AI-system position before screening the supported routes.
  • Records factual evidence for manipulation, vulnerability, social scoring, criminal prediction, biometric and emotion routes.
  • Stops the preflight when a potential prohibited-practice indicator is recorded and keeps uncertain scope not assessable.

From official material to a working record

Official material sets the basis

The EU AI Act and Commission guidance define prohibited-practice routes and explain their interpretation.

The tool prepares the operational record

The preflight turns those routes into a conservative factual record with evidence, owners and a clear stop condition.

Official sources stay visible

Each supported check shows its jurisdiction, source title, source version and review date beside the result. Unsupported cases are rejected rather than estimated.

View this product's official sources

Questions before you use the tool

Scope, files, evidence and browser-local handling.

Does a no-indicator result mean the use is lawful?

No. It records only that no supported indicator was identified in the facts entered and reviewed.

What happens if scope is uncertain?

The assessment fails closed. Items remain not assessable until a responsible person confirms the EU connection and AI-system position.

Does the tool assess exceptions?

It records an exception or authorisation reference, but a specialist must determine whether that route is available and satisfied.

Where are the use-case facts processed?

The entered facts and generated screening files remain in this browser on this device.

Screen the proposed AI use

Confirm the supported route, record the operating facts and inspect each potential indicator.

Open the preflight