FAQEU

EU AI Act Who needs a rights impact assessment?

Certain deployers must assess how their planned high-risk AI use could affect people's fundamental rights. Article 27 calls this a fundamental rights impact assessment, or .

Check both the Annex III use case and the deployer category before first use. Critical-infrastructure systems under Annex III point 2 are excluded from this duty.

Author
Sorena AI
Published
May 9, 2026
Updated
Oct 21, 2026
Overview

A is required for specified public-service, credit and insurance uses of Annex III high-risk AI. It records who may be harmed, how people will oversee the system and what happens if risks occur. The duty applies from 2 December 2027 under the 2026 amendment, published on 24 July 2026 and in force from 27 July. First check the exact system and deployer conditions below.

Search this module

Find a question or answer quickly

3 of 3 questions
Question 1

Who must complete a FRIA?

Start with the system: it must be high-risk under Article 6(2), the Annex III use-case route. Article 27 excludes the critical-infrastructure systems listed in Annex III point 2.

Then check the deployer. The duty covers bodies governed by public law, private entities providing public services, and deployers of Annex III point 5(b) creditworthiness or credit-scoring systems or point 5(c) life and health insurance risk-assessment and pricing systems.

Complete the assessment before first use. You may rely on an earlier or provider assessment in a similar case. Check that the system, process, affected groups, risks, oversight and measures fit your deployment. Update the information if an element changes or becomes outdated during use.

  • Start with Article 6(2): confirm that the system is an Annex III high-risk AI system.
  • Check the carve-out: Annex III point 2 critical-infrastructure systems are excluded from Article 27 , even though they may still be high-risk and are registered at national level under Article 49(5).
  • Check the deployer category: public-law bodies, private entities providing public services, and deployers using Annex III point 5(b) or 5(c) systems are the Article 27 categories.
  • Do not treat a provider's high-risk classification memo as a ; Article 27 is a deployer-side assessment of the specific use.
  • After completing the , notify the of the results by submitting the completed Article 27 template, unless the Article 46(1) exemption from notification applies.

Does every high-risk system need a ?

No. Article 27 applies to specified deployers before deploying Article 6(2) Annex III high-risk systems, with an express exception for the Annex III point 2 critical-infrastructure area. Product-safety high-risk systems classified under Article 6(1), and Annex III systems outside the named deployer categories, should still be assessed for other AI Act duties, but Article 27 is not automatically triggered by the high-risk label alone.

Which deployers are covered?

Article 27 names deployers that are bodies governed by public law, private entities providing public services, and deployers of high-risk systems referred to in Annex III points 5(b) and 5(c). Recital 96 explains that private public-service examples can be linked to public-interest tasks such as education, healthcare, social services, housing, and administration of justice.

What about critical-infrastructure systems?

Article 27 excludes high-risk AI systems intended for the Annex III point 2 critical-infrastructure area from the duty. That does not remove all AI Act obligations: Annex III point 2 covers safety components in critical digital infrastructure, road traffic, and water, gas, heating, or electricity supply, and Article 49(5) says those high-risk systems are registered at national level.

When does the duty apply?

Regulation (EU) 2026/1744 moves Chapter III Sections 1 to 3, except Article 6(5), to 2 December 2027 for Article 6(2) Annex III high-risk systems. This includes Article 27. For systems placed on the market or put into service before that applicable date, amended Article 111(2) separately governs transition treatment, including significant design changes and the 2 August 2030 deadline for high-risk systems intended for public-authority use. It preserves Article 5 prohibitions; it is not a grace period for every new deployment.

Question 2

What must the assessment cover?

Describe the process in which you will use the system, how long and how often it will be used, and the people or groups likely to be affected. Assess the specific risks of harm to those groups, considering the provider's Article 13 information.

Explain how people will oversee the system according to its instructions and how you will respond if risks occur. Include internal governance and complaint mechanisms. Connect the assessment to the operating procedures people will follow.

  • Process evidence: workflow map, intended purpose, provider instructions for use, and the deployer's use case.
  • Use evidence: expected start, duration or period of use, frequency, countries or operating units, and whether this is a first use or a similar case relying on an earlier assessment.
  • Affected-person evidence: natural-person categories, affected groups, dependency or vulnerability factors, and the decision or service the AI output influences.
  • Risk evidence: specific fundamental-rights harms, provider Article 13 information considered, residual risks, and escalation criteria.
  • Control evidence: human oversight procedure, complaint route, internal governance owner, operational playbook for risk materialisation, and approval record.

Can we reuse an earlier assessment?

Yes, but only in similar cases. Article 27 says the obligation applies to the first use of the high-risk AI system and that a deployer may rely on previous FRIAs or existing impact assessments in similar cases. The record should explain why the earlier assessment is similar enough for the current process, affected groups, risks, oversight, and complaint arrangements.

When must we update it?

Article 27 requires an update when, during use, the deployer considers that any assessment element has changed or is no longer up to date. Practical update triggers include a changed deployment process, materially different use frequency, a new affected group, changed provider instructions, new risk evidence, or changed oversight, governance, or complaint mechanisms.

What records should we keep?

Keep the classification decision, the deployer-category check, the Annex III point check, provider instructions used for the risk analysis, the Article 27 assessment answers, evidence of human oversight and complaint routing, the market-surveillance notification record, and, where applicable, the link and EU database or national registration evidence.

Question 3

How do privacy assessment, notification and registration fit?

A data protection impact assessment () addresses personal-data risks under the General Data Protection Regulation (GDPR) or law-enforcement data-protection rules. A has a wider fundamental-rights scope. Under amended Article 27(4), you may cross-reference or include parts of a DPIA under GDPR Article 35 or Directive (EU) 2016/680 Article 27 that already meet a FRIA requirement. Complete every remaining Article 27 element.

After assessment, notify the by submitting the completed Article 27 template, unless the Article 46(1) notification exemption applies. Separately, covered public authorities, Union institutions, bodies, agencies and offices, and people acting for them, must register their Annex III system use before it starts. Annex III point 2 systems use national registration.

  • Map the and where personal-data risk and fundamental-rights risk overlap. Cross-reference or include the DPIA sections that meet Article 27 elements, then complete the remaining FRIA fields.
  • Keep the market-surveillance authority notification proof with the completed Article 27 template once the is performed.
  • For public-authority deployers, confirm Article 49 registration before putting the Annex III high-risk system into service or use.
  • For law enforcement, migration, asylum, and border-control Annex III systems, expect restricted EU database registration rules under Article 49(4).
  • For Annex III point 2 critical-infrastructure systems, route registration evidence to the national-level process described in Article 49(5).

Does a privacy impact assessment satisfy Article 27?

Not by itself. Under Article 27(4), as replaced by Regulation (EU) 2026/1744, the deployer may cross-reference relevant sections or include them in the when they already meet an Article 27 obligation. The deployer must still add any missing process, affected-group, fundamental-rights risk, human-oversight, governance, complaint, notification, and AI Act registration evidence.

Who receives the results?

Article 27 says the deployer must notify the of the results after the assessment is performed, using the filled-out template referred to in Article 27. Identify the applicable market surveillance authority for the Member State and deployment, and send the results to it. Article 27(3) permits an exemption from notification in the Article 46(1) derogation case; record whether that exemption applies.

What must be checked before first use?

Check that the system is an Article 6(2) Annex III high-risk system, the Annex III point 2 exclusion has been considered, the deployer category is documented, the Article 27 assessment fields are complete, the relationship is mapped where applicable, notification evidence is ready, and Article 49 registration evidence is present where the deployer is a public authority or acting on one.

Primary sources

References and citations

ai-act-service-desk.ec.europa.eu2 citations
Referenced sections
  • Commission-hosted AI Act Explorer page used to verify the public Article 27 page and official article heading.
  • Commission-hosted AI Act Explorer page for Article 27 used to cross-check the FRIA article citation and official article title.
"Article 27: Fundamental rights impact assessment for high-risk AI systems"
eur-lex.europa.eu
Referenced sections
  • Supports the required FRIA contents, the provider information needed to assess deployer-side risk, and the FRIA and DPIA summary fields required for applicable deployer registration.
"a description of the deployer’s processes"
eur-lex.europa.eu
Referenced sections
  • Supports the Article 27 trigger, covered deployer categories, critical-infrastructure carve-out, first-use rule, notification duty, DPIA complement rule, FRIA content list, and the related deployer registration requirements.
"Fundamental rights impact assessment for high-risk AI systems"
eur-lex.europa.eu
Referenced sections
  • Supports the FRIA notification, first-use, update, and DPIA complement rules; the registration obligations; and the distinction between product-safety and Annex III high-risk systems.
"the deployer shall notify the market surveillance authority of its results"
eur-lex.europa.eu
Referenced sections
  • Binding amendment published on 24 July 2026, entering into force on 27 July 2026, which replaces Article 27(4) and (5), permits DPIA cross-references or incorporation, and moves the Annex III Chapter III duties to 2 December 2027.
"include cross-references to the relevant sections"
Show all related guides (34)

What must general-purpose model providers do? | EU AI Act

Check general-purpose model duties, systemic-risk classification, open-source conditions, copyright, training summaries, incidents and application dates.

When must you disclose AI use? | EU AI Act

Check AI interaction notices, technical content marking, biometric notices, deepfake and text disclosures, exceptions, accessibility and deadlines.

When must you report a serious incident? | EU AI Act

Check serious-incident outcomes, reporting roles, 15-, 10- and two-day deadlines, authority recipients, corrective action and the separate model route.

Common questions, clear answers | EU AI Act

Clear answers on AI Act scope, risk categories, provider and user duties, general-purpose models, disclosures, rights, assessments and deadlines.

EU AI Act deadlines and compliance calendar

Find the AI Act deadline for your system or model. Check high-risk dates, transparency duties, older-model transitions and the evidence to prepare.

EU AI Act fundamental rights impact assessments

Check who needs an Article 27 fundamental rights impact assessment, what it must contain and how DPIA reuse, notification and registration work.

EU AI Act GPAI model evidence checklist

Collect model documentation, integration information, copyright records and training summaries. Add testing, incident and security records for systemic risk.

EU AI Act monitoring and serious incidents

Set up high-risk AI monitoring, identify serious incidents and check reporting deadlines, recipients, deployer escalation and corrective action.

EU AI Act penalties and fines

Check the AI Act fine tiers, smaller-organisation caps, GPAI provider fines and the facts authorities consider when deciding a penalty.

EU AI Act serious incident reporting steps

Check serious-incident harm, preserve evidence and choose the reporting deadline and recipient for high-risk AI systems and systemic-risk GPAI models.

EU AI Act timeline and phased deadlines

See which AI Act rules apply now and what comes next. Check high-risk dates, older-model transitions and the difference between legal and standards dates.

EU AI Act vs ISO/IEC 42001

Compare AI Act legal duties with ISO/IEC 42001 management-system requirements. See which records can be reused and what certification does and does not cover.

EU AI Act vs NIST AI RMF

Compare AI Act legal duties with the voluntary NIST AI RMF. Map risk records to specific requirements and identify legal steps that need separate evidence.

EU AI Act: Check whether the Act applies

Use five checks to establish AI Act scope, exclusions, your role, risk categories and evidence for the decision.

EU AI Act: Choose a high-risk assessment route

Choose between provider-run internal control, notified-body assessment and a product-law procedure before release.

EU AI Act: Classify a tool before approval

Classify an AI tool or model before purchase or release. Check EU scope, prohibited uses, high-risk rules, transparency and GPAI duties.

EU AI Act: Compliance checklist

Check each AI tool, decide which duties apply, and give every action an owner and evidence link.

EU AI Act: Conformity assessment

Some high-risk systems use a provider-run assessment. Others need an authorised independent assessor or a product-law assessment.

EU AI Act: Does it apply to you?

Check whether the AI Act applies to your system or model, identify your provider or deployer role, and keep evidence for the decision.

EU AI Act: Duties for general-purpose AI models

Check GPAI provider duties for documentation, integrator information, copyright and training summaries, plus extra requirements for systemic-risk models.

EU AI Act: High-risk requirements

High-risk providers must meet system requirements before supply. Deployers must use and monitor the system under the provider's instructions.

EU AI Act: High-risk system checklist

Check the system's risk controls, data, documentation, logs, instructions, human oversight and testing before release.

EU AI Act: Is your use case high-risk?

Find high-risk AI examples in recruitment, banking, education, healthcare, public services and products. Check Article 6 conditions and exceptions.

EU AI Act: Plan your compliance work

Build a compliance plan for each AI tool: your role, the duties that apply, the evidence needed and the person responsible.

EU AI Act: Standards and quality management

Use EN 18286:2026 to organise provider quality-management procedures and keep evidence that they work for each high-risk system.

EU AI Act: Technical file and evidence templates

Build the technical file and provider records for a high-risk AI system, with owners and evidence links for each field.

EU AI Act: When must you disclose AI?

Check Article 50 chatbot notices, synthetic-output marking, biometric notices, deepfake disclosures and public-interest text exceptions.

EU AI Act: Which AI uses are banned?

Check Article 5 prohibited AI practices, the narrow exceptions, evidence to retain and the new content prohibitions from 2 December 2026.