Do we have to keep logs of our AI system?

education

Logging in admission and assessment: what the institution keeps in its own hands

A university of applied sciences uses a purchased AI system that ranks student admissions and also raises flags during digital assessment. The logs sit in the supplier environment, which hands them over on request. The teaching organisation wonders whether that settles the matter or whether the school retains a duty of its own.

Provenance: Article 12(1) requires a high-risk AI system to technically allow for the automatic recording of events (logs) over the lifetime of the system, and Article 12(2) ties that recording to a level of traceability appropriate to the intended purpose of the system. Article 26(6) provides that deployers keep the automatically generated logs to the extent those logs are under their control, for a period appropriate to the intended purpose, of at least six months, unless applicable Union or national law provides otherwise, in particular Union law on the protection of personal data. Article 19(1) places a corresponding retention duty on providers for the logs generated by their systems that are under their control, likewise for at least six months.

On our reading, the fact that the supplier hosts the logs does not by itself place the school outside Article 26(6): what matters is whether the logs are under its control. The Regulation does not define that notion, and on our reading server location is not decisive in itself; the question is whether you can actually and contractually dispose of those logs, and it has to be answered for each procurement. So agree at procurement that the institution can request, export and itself retain the logs for the chosen period, and record which admission and assessment events that covers. Take data protection into account at the same time, because the same provision allows other Union or national law to cap the retention period.

Editorial example. The rule above is in the Regulation. The situation was written by us to show how that rule plays out in this sector, and is not taken from a worked case in official guidance.

Article 12(1) and (2), Article 19(1) and Article 26(6)

manufacturing and industry

Logging on the production line: which logs the manufacturer keeps and which the factory keeps

A manufacturer supplies an AI system that runs as a safety component inside the machinery of a production line and also drives quality control. The factory operating the line keeps only the alerts visible in the local controller; the rest of the recording flows to the supplier environment. The question is who has to keep which logs when it later has to be reconstructed why the line was halted.

Provenance: Article 12(1) requires high-risk AI systems to technically allow for the automatic recording of events (logs) over the lifetime of the system. Under Article 12(2), those logging capabilities must enable the recording of events relevant for identifying situations that may result in the system presenting a risk within the meaning of Article 79(1) or in a substantial modification, for facilitating the post-market monitoring referred to in Article 72, and for monitoring the operation of the high-risk AI systems referred to in Article 26(5). Article 19(1) obliges providers to keep those automatically generated logs to the extent they are under their control, for a period appropriate to the intended purpose of the system, of at least six months, unless applicable Union or national law provides otherwise. Article 26(6) places a corresponding retention duty on deployers for the logs under their control, with the same six-month floor.

We read Article 19 together with Article 26(6) as leaving the manufacturer, acting here as the provider, and the factory, acting as the deployer, each with a retention duty of its own for the logs under its control. The Regulation does not say when logs count as being under your control, and on our reading the place where the recording physically lands does not settle that by itself; the question has to be answered system by system. For a production line that means recording which events stay in the machinery controller and which travel to the supplier, and securing access to that second set contractually before you need it. Leave that unarranged and you risk being unable to trace a line stoppage or a quality control rejection back to the behaviour of the system.

Editorial example. The rule above is in the Regulation. The situation was written by us to show how that rule plays out in this sector, and is not taken from a worked case in official guidance.

Article 12(1) and (2), Article 19(1) and Article 26(6)

workplace and staff

Logging in task allocation at work: evidence about the system or a file on the employee

An employer deploys an AI system that handles task allocation among staff and summarises their performance for the performance review. HR wants to know what record of those outcomes has to be retained when an employee objects months later to a promotion decision.

Provenance: Article 12(1) requires high-risk AI systems to technically allow for the automatic recording of events (logs) over the lifetime of the system. Article 12(2)(c) names, as one of the purposes of those logging capabilities, monitoring the operation of the systems referred to in Article 26(5), and that paragraph obliges deployers to monitor operation on the basis of the instructions for use and, where relevant, to inform the provider in accordance with Article 72. Article 26(6) obliges deployers to keep the automatically generated logs to the extent they are under their control, for a period appropriate to the intended purpose, of at least six months, unless applicable Union or national law provides otherwise. Article 19(1) sets out a corresponding retention duty for providers, with the same six-month floor.

We read Article 12(2) as putting the logging capabilities there first of all to follow risks, substantial modifications and the operation of the system, not to sharpen judgements about individual staff. Whether that statement of purpose also limits what the retained logs may later be used for is something Article 12 does not say: on our reading that limit has to come from data protection law, to which Article 26(6) itself refers. For HR a practical line follows: keep what is needed to trace an outcome and to carry out the monitoring under Article 26(5), and settle in writing beforehand whether those same files may double as a performance record. Bear in mind that the six-month floor can be shorter than the period within which an employee challenges a promotion decision; the Regulation does not govern that evidential position.

Editorial example. The rule above is in the Regulation. The situation was written by us to show how that rule plays out in this sector, and is not taken from a worked case in official guidance.

Article 12(1) and (2)(c), Article 19(1) and Article 26(5) and (6)

recruitment and selection

Candidate recommendation that automatically becomes a decision

An employer uses a system that ranks applicants and recommends a candidate to hire. In one setup a recruiter weighs that recommendation in their own assessment; in the other the outcome is applied automatically and a candidate is rejected without anyone looking at it.

Provenance: The Commission draft guidelines of 19 May 2026 address this case when determining whether an application falls under Annex III. The document is a consultation version: non-binding and not yet final.

Assess a recruitment system on its intended purpose rather than on whether a recruiter reviews the output, because adding or removing human involvement does not change its high-risk classification.

Draft guidelines on high-risk AI classification, 19 May 2026, annex on Annex III

Article 6 has two separate routes to high-risk

The European Commission's draft guidelines on the classification of high-risk AI of 19 May 2026, which are expressly non-binding, state in paragraph (7) that an AI system is high-risk in two scenarios: first, where it is intended to be used as a safety component of a product, or is itself a product, covered by the Union harmonisation legislation listed in Annex I and required to undergo third-party conformity assessment; and second, where it falls within one of the use cases in the areas listed in Annex III. Paragraph (448) of those same draft guidelines notes that the Article 113 application dates have been postponed by the AI Omnibus to 2 December 2027 for the Article 6(2) route and 2 August 2028 for the Article 6(1) route.

Draft guidelines on high-risk AI classification (19 May 2026), General principles chapter, section II, paragraph (7); section V, paragraph (448)

Broadly positioned and general purpose AI systems: a disclaimer is not enough

According to the non-binding draft guidelines of 19 May 2026 on the classification of high-risk AI, paragraph (12) provides that where the instructions for use, contractual arrangements, terms of service, usage policy, promotional and sales materials or technical documentation present the AI system as broadly applicable across a generality of contexts and functions, and do not consistently limit its application or exclude high-risk uses, the system's intended purpose will be deemed to also encompass high-risk use cases and therefore qualify as high-risk. Under these draft guidelines this applies in particular where such uses are feasible and reasonably foreseeable given the system's functionalities and capabilities. The same paragraph states that merely asserting, for example in the terms of service, that high-risk uses are excluded is insufficient where the provider's overall presentation, examples or product positioning effectively provides for or promotes such uses, and that any limitations of use must be described clearly, concretely and coherently across all materials.

Draft guidelines on high-risk AI classification (19 May 2026), General principles chapter, section II.2, paragraph (12)

High-risk does not mean prohibited, and not high-risk does not mean permitted

The draft guidelines of 19 May 2026 on the classification of high-risk AI, which are not binding, state in paragraph (3) that the fact an AI system is listed as an example in these guidelines does not mean its use should automatically be considered lawful, since such use would still need to comply with other applicable legislation, and in paragraph (4) that the scope of these guidelines is limited to whether an AI system is high-risk or not. In the Annex III chapter of this draft, paragraph (68) states that classifying systems as high-risk under Article 6(2) does not mean their use is prohibited, but that those systems are subject to appropriate requirements. Paragraphs (82) and (83) of this draft explain the wording in so far as their use is permitted under relevant Union or national law and state that falling within a use case does not necessarily mean the system may lawfully be used in those cases, that in addition to the prohibitions other provisions of Union or national law may restrict use, and that under Article 2(9) the AI Act applies without prejudice to rules on consumer protection, product safety and data protection.

Draft guidelines on high-risk AI classification (19 May 2026), General principles chapter, paragraphs (3) and (4); Annex III chapter, paragraph (68) and section 2.6, paragraphs (82) and (83)

Split and agentic architectures are assessed as a whole

The non-binding draft guidelines of 19 May 2026 provide in paragraphs 75, 76 and 90 that where several AI systems form part of a more complex whole and their combined intended purpose or joint outputs materially influence an individual decision, that configuration is treated as a single AI system for classification. The draft expressly states that split architectures are assessed as a whole to prevent circumvention by system design, that exemptions for individual modules do not apply where the overall configuration influences key aspects of the decision, and that this also extends to complex interconnected setups such as agentic AI systems whose linked actions jointly serve a high-risk purpose. Under the same draft, strictly procedural or preparatory functions do remain eligible for exemption where they are genuinely separable from the system and do not structure or feed outputs that materially influence the examination of an individual case.

Section IV.2.3, paragraphs 75 and 76, and section IV.2.7.1 paragraph 90

prEN 18229-1: AI trustworthiness framework part 1, logging

prEN 18229-1 (AI trustworthiness framework, Part 1: Logging) is the JTC 21 deliverable under M/613 for Article 12 of the AI Act: high-risk AI systems must technically allow for the automatic recording of events over their lifetime. As at June 2026 the deliverable was at the Enquiry stage. It has not yet been published as an EN and is not cited in the Official Journal.