Point 7(c) of Annex III classifies as high-risk AI systems intended to be used by or on behalf of competent public authorities or by Union institutions, bodies, offices or agencies to assist competent public authorities in the examination of applications for asylum, visa or residence permits and with associated complaints as regards the eligibility of the natural persons applying for a status, including related assessments of the reliability of evidence. Such systems are classified as high-risk because they have the potential to materially influence the decisions of public authorities that affect people who are often in a particularly vulnerable position and errors, bias, or opacity may lead to severe, and sometimes irreversible harm (e.g., due to refusal of status or exposure to removal).
Point 7(c) covers AI systems intended to assist competent authorities in examining the eligibility of applicants for asylum, visa or residence permits and in examining associated complaints that relate to the eligibility determination. This includes compla ints concerning elements capable of affecting either the substantive assessment of the application (such as facts, evidence, or credibility) or the procedural fairness of the decision-making process (such as the right to be heard, interpretation, or reason ing). It does not cover complaints concerning general service quality, understood as complaints about organisational or logistical aspects of service delivery that do not affect the substance of the eligibility assessment or the procedural fairness of the decision-making process.
The elements listed in point 7(c) of Annex III are alternative (asylum or visa or residence; applications or associated complaints), not cumulative. The terms ‘including related assessment of the reliability of evidence’ clarifies that reliabilityof-evidence checks are illustrative, not-exhaustive. The concept of ‘assistance’ is not limited to the decisions e.g. on application for international protection. It also covers earlier stages of the examination (e.g. steps that analyse interview content, documents or country-of-origin information and are relevant for the decision on the merits. Where the AI system’s outputs materially influence the substance or outcome of eligibility determinations, such as by validating or challenging evidence, highlighting inconsistencies, prioritising issues, or pre-classifying status elements it should be classified as high-risk pursuant to point 7(c) of Annex III. For example, if an AI system derives its outputs from the substance of a live case (statements, documents, COI) and uses them to prioritise, flag, validate or challenge evidence, or to recommend next steps, it materially influences the examination and remains within point 7(c).While the filter mechanism in Article 6(3) may be relied upon for narrow procedural or purely preparatory tasks, the systems assessing reliability of evidence generally materially influence decision-making and should in principle be classified as high-risk (e.g. in case of detection of document morphing).
Where a single integrated assistant is intended to assist in the examination of eligibility (and associated complaints) and its outputs can materially influence substance (e.g. by validating or challenging evidence, prioritising issues, or pre-classifying status elements), the overall AI system falls within the use case listed in point 7(c) of Annex III. If individual functionalities are truly separable and put into service independently (e.g. as stand-alone tools), each functionality may be assessed on its own for possible high-risk classification. Some modules can qualify for as the exceptions for AI systems intended to perform narrow procedural or purely preparatory listed in Article 6(3)(a) and (d) AI Act, while others remain high-risk. However, low-risk modules inside a single assistant that steers eligibility analysis, cannot benefit from the filter mechanism. Nor does embedding low-risk modules inside a single assistant that steers eligibility analysis bring the overall system under the filter mechanism. For the exception in Article 6(3)(c) AI Act to apply, the system must be intended to perform ex-post detection of patterns or deviations.
a) Practical examples of AI systems falling within the use case of point 7(c) of Annex III
Evidence appraisal and consistency analysis
- An AI system intended to assess the authenticity and consistency of documents or the plausibility of personal narratives and to provide credibility signals that the examiner relies on when determining eligibility of migrants to entry the country or asylum-seekers. These outputs validate or challenge evidence relied upon by the examiner and therefore assist the examination.
- An AI system intended to analyse phone data lawfully obtained from a device to validate routes, timelines or contacts and to produce signals used in an eligibility appraisal. This assists the authority’s appraisal of evidence in the file.
- An AI system intended to detect document morphing, non-conforming inks, copied signatures or chip and public-key infrastructure validity as part of document verification as a step in the examination procedure for assessing eligibility for a visa, a residence permit or international protection. Automated comparison of discrepancies against prior submissions
- An AI system intended to identify substantive or credibility-relevant discrepancies against prior submissions in order to assist the authority’ssubsequent examination of the application for asylum, a visa, a residence permit, or an associated complaint. Such an AI system assists the examination. Origin inference used in the examination
- An AI system intended to analyse recorded speech to indicate a likely country or region of origin for use in the authority’s eligibility reasoning and selection of countryof-origin-information. The output assists the examiner when deciding whether information is treated as proven, contested, prioritised or further investigated in a specific person’s file.
b) Practical examples of an AI system falling outside the use case of point 7(c) of Annex
- An AI system intended to provide applicants with answers to frequently asked questions, to guide users to the correct forms, to schedule appointments, or to deliver application status notifications. Such an AI system does not assist the authority in examin ing eligibility or associated complaints, but its intended purpose is service delivery to migrant applicants.
c) Practical examples of AI systems falling within the use case but exempted under the filter mechanism of Article 6(3)
Narrow procedural tasks (Article 6(3)(a) AI Act)
- An AI system intended only to organise a case file by applying pre-defined categories for storage and retrieval (for example, ‘identity documents”, ‘medical records”, ‘interview notes”), without ranking materials by importance or omitting any items. Applying fixed, predefined labels for storage and retrieval is a narrow procedural operation. The system does not evaluate evidence, set priorities, or influence the assessment.
- An AI system intended to check the completeness and formatting of visa applications, to sort or assign files for administrative handling based on public and objective criteria, while performing only technical and non-interpretative operations and without assessing the content of the application, determining the applicable procedural track, or otherwise steering the substance of the examination.
- An AI system intended to highlight verbatim differences between current and previous statements (for example, string-level contrasts), without generating any credibility assessment or recommending follow-up questions; where the system evaluates the significance of the differences or labels them as credibility concerns, the exception no longer applies.
- A pre-check tool that, under strict, objective rules (e.g., a visa facilitation agreement fixes the fee for applicants from a specific country), routes an applicant to payment of a fixed fee.
- AI system intended to check the completeness or format of travel authorisation or visa (ETIAS/VIS) files (e.g. required fields, document presence), without producing any risk score, flag or triage of individuals. Such a system performs narrow procedural tasks, rather than preparatory tasks, since it does not prepare or structure substantive content for the examination. Preparatory tasks (Article 6(3)(d) AI Act)
- An AI system intended to transcribe interviews, translate interview transcripts or produce neutral summaries for the file, without advising on next steps or suggesting outcomes. A system that goes beyond these functions, for example by introducing elements not contained in the applicant’s statements, making credibility assessments, or l inking the applicant’s statements to the applicable legal framework, will not benefit from the exception.
- An AI system that carries out large data scanning on electronic devices to extract metadata on the travel route taken by an individual. The system sets out a neutral output from the search providing all pieces of metadata relating to geolocation without preferring one piece of information over another. As such, it could be considered to carry out a preparatory task. Improving a completed human activity (Article 6(3)(b))
- An AI system intended to harmonise terminology and correct grammar in a decision after the human assessment has been concluded, without introducing additional reasoning or altering the outcome. A system that infers or formulates legal or factual grounds no t explicitly stated or re-evaluates credibility is not covered by the exception. Ex-post pattern detection (Article 6(3)(c))
- An AI system intended to analyse past, completed eligibility files in order to detect decision-making patterns or deviations for quality-assurance reporting, without proposing outcomes in actual cases.