{"product_id":"ai-act-50-2-maskinlasbar-markning-detektion-testbevis","title":"AI Act Article 50.2 – machine-readable AI labelling, detection and proof of testing | Word\/PDF\/Excel","description":"\n\u003cp\u003e\u003cstrong\u003eDoes your company develop or provide an AI system that generates synthetic text, images, audio, or video?\u003c\/strong\u003e This Swedish template package helps affected AI providers document \u003cstrong\u003emachine-readable labelling, actual detectability, technical testing, and release decisions\u003c\/strong\u003e in accordance with Article 50.2 of the EU AI Act. The package contains working documents and registers – it does not install any watermarking, detector, or certified technology.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003ePurchase overview:\u003c\/strong\u003e 7 independent documents (five Swedish working templates, a Swedish user guide, and a separate English User Guide) as editable Word files and corresponding searchable, printable PDF files; plus a nine-tab Excel workbook in Swedish. A total of \u003cstrong\u003e15 customer files\u003c\/strong\u003e in a designated ZIP package for digital download after future activation. The PDF files are \u003cstrong\u003enot interactively fillable\u003c\/strong\u003e. The main documents are in Swedish; the English guide is an instruction, not a legally translated main template. Swedish usage context and EU law, version 1.0 (2026-11-09).\u003c\/p\u003e\n\u003cdiv style=\"border:2px solid #27628d;padding:15px;margin:16px 0;border-radius:8px;background:#f4f9fd;\"\u003e\n\n\u003ch2\u003e01. Package contents\u003c\/h2\u003e\n\n\u003cp\u003e\u003cstrong\u003e7 independent documents:\u003c\/strong\u003e Applicability assessment, technical specification, detection test protocol, non-conformance and corrective action protocol, release and evidence decision, Swedish guide, and English User Guide. All in DOCX and searchable PDF (14 format files). In addition, \u003cstrong\u003e1 XLSX\u003c\/strong\u003e with nine tabs. A total of 15 customer files. PDF copies are not additional independent documents.\u003c\/p\u003e\n\n\n\u003c\/div\u003e\n\u003cdiv style=\"border:2px solid #27628d;padding:15px;margin:16px 0;border-radius:8px;background:#f4f9fd;\"\u003e\n\n\u003ch2\u003e02. What the files contain\u003c\/h2\u003e\n\n\u003cp\u003eRole and exemption assessment under the AI Act, selection of labelling method and detection tool, actual testing of text\/image\/audio\/video, robustness after file conversion, non-conformance tracking, retesting, release decision, and traceable source references. Excel calculates \u003cstrong\u003eadministrative counts and statuses\u003c\/strong\u003e, but does not automatically assess legal applicability or technical compliance.\u003c\/p\u003e\n\n\n\u003c\/div\u003e\n\u003ch2\u003eWho should choose this package?\u003c\/h2\u003e\n\u003cp\u003e\u003cstrong\u003eExample A:\u003c\/strong\u003e A Swedish software provider builds its own image-generating AI system and needs to show how the output is labelled and how the labelling can actually be detected after export. \u003cstrong\u003eExample B:\u003c\/strong\u003e A provider of generative text or audio needs to document its role, technical exemptions, test results, and the next release before deployment.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eDo not choose this as your only solution\u003c\/strong\u003e if the company only uses an external AI tool as a deployer, if the customer wants to label deepfakes under Article 50.4, or if a full conformity assessment of a high-risk AI system is required. The product's forms do not replace actual implementation, external legal assessment, or a competent authority's decision.\u003c\/p\u003e\n\u003ch2\u003eFile by file – concrete practical utility\u003c\/h2\u003e\n\u003cdiv style=\"overflow-x:auto\"\u003e\u003ctable\u003e\n\n\u003cthead\u003e\u003ctr\u003e\n\n\u003cth scope=\"col\"\u003eDocument\u003c\/th\u003e\n\n\u003cth scope=\"col\"\u003eWhen it is used\u003c\/th\u003e\n\n\u003cth scope=\"col\"\u003eLanguage\u003c\/th\u003e\n\n\u003cth scope=\"col\"\u003eFormat\u003c\/th\u003e\n\n\n\u003c\/tr\u003e\u003c\/thead\u003e\n\n\u003ctbody\u003e\n\n\u003ctr\u003e\n\n\u003ctd\u003e01 Applicability and provider role\u003c\/td\u003e\n\n\u003ctd\u003eAssess system, actor role, output, exemptions, and transition period\u003c\/td\u003e\n\n\u003ctd\u003eSwedish\u003c\/td\u003e\n\n\u003ctd\u003eDOCX + PDF\u003c\/td\u003e\n\n\n\u003c\/tr\u003e\n\n\u003ctr\u003e\n\n\u003ctd\u003e02 Technical labelling specification\u003c\/td\u003e\n\n\u003ctd\u003eDocument labelling, detector, format, limitations, and owner\u003c\/td\u003e\n\n\u003ctd\u003eSwedish\u003c\/td\u003e\n\n\u003ctd\u003eDOCX + PDF\u003c\/td\u003e\n\n\n\u003c\/tr\u003e\n\n\u003ctr\u003e\n\n\u003ctd\u003e03 Technical detection test\u003c\/td\u003e\n\n\u003ctd\u003eTesting of original file, export, transcoding, and negative controls\u003c\/td\u003e\n\n\u003ctd\u003eSwedish\u003c\/td\u003e\n\n\u003ctd\u003eDOCX + PDF\u003c\/td\u003e\n\n\n\u003c\/tr\u003e\n\n\u003ctr\u003e\n\n\u003ctd\u003e04 Non-conformance and correction\u003c\/td\u003e\n\n\u003ctd\u003eLog errors, temporary measure, root cause investigation, and retesting\u003c\/td\u003e\n\n\u003ctd\u003eSwedish\u003c\/td\u003e\n\n\u003ctd\u003eDOCX + PDF\u003c\/td\u003e\n\n\n\u003c\/tr\u003e\n\n\u003ctr\u003e\n\n\u003ctd\u003e05 Release and evidence handover\u003c\/td\u003e\n\n\u003ctd\u003eDefined internal decision and references to approved test\u003c\/td\u003e\n\n\u003ctd\u003eSwedish\u003c\/td\u003e\n\n\u003ctd\u003eDOCX + PDF\u003c\/td\u003e\n\n\n\u003c\/tr\u003e\n\n\u003ctr\u003e\n\n\u003ctd\u003e06 Swedish user guide\u003c\/td\u003e\n\n\u003ctd\u003eSteps, legal boundaries, common misunderstandings, and sources\u003c\/td\u003e\n\n\u003ctd\u003eSwedish\u003c\/td\u003e\n\n\u003ctd\u003eDOCX + PDF\u003c\/td\u003e\n\n\n\u003c\/tr\u003e\n\n\u003ctr\u003e\n\n\u003ctd\u003e07 English User Guide\u003c\/td\u003e\n\n\u003ctd\u003eInstructions for completing the Swedish templates\u003c\/td\u003e\n\n\u003ctd\u003eEnglish\u003c\/td\u003e\n\n\u003ctd\u003eDOCX + PDF\u003c\/td\u003e\n\n\n\u003c\/tr\u003e\n\n\u003ctr\u003e\n\n\u003ctd\u003e08 Labelling and test register\u003c\/td\u003e\n\n\u003ctd\u003eSystem, output, labelling, tests, non-conformances, decisions, and evidence\u003c\/td\u003e\n\n\u003ctd\u003eSwedish\u003c\/td\u003e\n\n\u003ctd\u003eXLSX\u003c\/td\u003e\n\n\n\u003c\/tr\u003e\n\n\n\u003c\/tbody\u003e\n\n\n\u003c\/table\u003e\u003c\/div\u003e\n\u003ch2\u003eThe Excel tool – what has been tested?\u003c\/h2\u003e\n\u003cp\u003eThe workbook has nine tabs: \u003cstrong\u003eOverview, System and Roles, Output and Scope, Labelling Technology, Detection Tests, Non-conformances, Decisions and Release, Evidence Register, and Read Me\u003c\/strong\u003e. The overview counts registered IDs and explicit statuses. A fictitious test with a technical result of “Failed” and a release status of “For assessment” has been tested and subsequently removed. A passed technical test is \u003cem\u003enot\u003c\/em\u003e automatically a legal approval. No formula designates a product as AI Act-compliant.\u003c\/p\u003e\n\u003ch2\u003eWorkflow: from rule assessment to release decision\u003c\/h2\u003e\n\u003col\u003e\n\n\u003cli\u003eDetermine who is the provider of the AI system and what types of synthetic output are affected.\u003c\/li\u003e\n\n\u003cli\u003eAssess whether Article 50.2 applies, and separately document possible exemptions and transitional rules.\u003c\/li\u003e\n\n\u003cli\u003eDescribe the actual machine-readable labelling, how it is implemented, and which detector is used.\u003c\/li\u003e\n\n\u003cli\u003eTest output in relevant formats and distribution chains. Save the sample files' IDs, results, and limitations.\u003c\/li\u003e\n\n\u003cli\u003eRecord failed tests, correct them, and perform retesting.\u003c\/li\u003e\n\n\u003cli\u003eDecide internally on the exact version, conditions, risks, and archiving. An internal signature is not an authority certification.\u003c\/li\u003e\n\n\n\u003c\/ol\u003e\n\u003ch2\u003eLegal and time limits – important differences\u003c\/h2\u003e\n\u003cp\u003e\u003cstrong\u003eArticle 50.2:\u003c\/strong\u003e Providers of affected AI systems shall ensure machine-readable labelling and that synthetic content can be detected, with requirements for effective, interoperable, robust, and reliable solutions \u003cem\u003eto the extent technically possible\u003c\/em\u003e. There are exemptions, including for certain standard editing functions and output that does not substantially alter the input or semantics. If an exemption applies, it must be assessed for each actual system. See \u003ca href=\"https:\/\/%20ai-act-service-desk.ec.europa.eu\/en\/ai-act\/article-50\" rel=\"noopener\"\u003eAI Act Article 50.2 – EU AI Act Service Desk\u003c\/a\u003e.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eArticle 111.4:\u003c\/strong\u003e There is a limited transitional period until \u003cstrong\u003e2 December 2026\u003c\/strong\u003e for affected systems placed on the market before \u003cstrong\u003e2 August 2026\u003c\/strong\u003e. It is not a general reprieve for all AI systems. See \u003ca href=\"https:\/\/ai-act-service-desk.ec.europa.eu\/en\/ai-act\/article-111\" rel=\"noopener\"\u003eEU Regulation Article 111.4\u003c\/a\u003e.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eTechnical methods:\u003c\/strong\u003e Watermarking, metadata, and cryptographic provenance can be alternative technical methods depending on the type of content; the law does not generally prescribe that a single particular standard must be used. Read the \u003ca href=\"https:\/\/digital-strategy.ec.europa.eu\/en\/policies\/guidelines-ai-transparency-obligations\" rel=\"noopener\"\u003eEU Commission guidelines\u003c\/a\u003e. The \u003ca href=\"https:\/\/digital-strategy.ec.europa.eu\/en\/faqs\/code-practice-transparency-ai-generated-content\" rel=\"noopener\"\u003eEU Code of Practice\u003c\/a\u003e is voluntary and does not replace the actual legal obligation.\u003c\/p\u003e\n\u003ch2\u003eCompare with related documents\u003c\/h2\u003e\n\u003cp\u003eThe \u003ca href=\"https:\/\/mallbutiken.se\/products\/ai-governance-paket-ai-act-compliance-2026\"\u003eAI Governance Package\u003c\/a\u003e is a broader foundation for AI governance. \u003ca href=\"https:\/\/mallbutiken.se\/products\/ai-deepfake-ai-influencer-markning-artikel-50\"\u003eDeepfakes and AI influencers under Article 50.4\u003c\/a\u003e primarily concerns a different obligation, mainly related to the provider's disclosures. This product focuses specifically on the \u003cstrong\u003eprovider's technical labelling and detectability\u003c\/strong\u003e under Article 50.2.\u003c\/p\u003e\n\u003ch2\u003eFrequently Asked Questions\u003c\/h2\u003e\n\u003ch3\u003eIs Article 50.2 the same as deepfake labelling?\u003c\/h3\u003e\u003cp\u003e\u003cstrong\u003eNo.\u003c\/strong\u003e Article 50.2 is aimed at the provider's machine-readable labelling and detection. Article 50.4 addresses separate disclosure obligations for certain uses, including deepfakes. An actor may be affected by different obligations depending on its role.\u003c\/p\u003e\n\u003ch3\u003eDo we have to use C2PA or watermarking?\u003c\/h3\u003e\u003cp\u003e\u003cstrong\u003eNot automatically a particular named method.\u003c\/strong\u003e The actual solution must be assessed against Article 50.2, current guidance, content type, and technical feasibility. The template helps you document choices and tests, but does not provide the technology itself.\u003c\/p\u003e\n\u003ch3\u003eCan an ordinary AI user use the template?\u003c\/h3\u003e\u003cp\u003e\u003cstrong\u003eOnly if the actual need is provider responsibility and technical testing.\u003c\/strong\u003e Companies that only use external tools should first determine their legal role and may need other documents.\u003c\/p\u003e\n\u003ch3\u003eWhat does “Passed” mean in Excel?\u003c\/h3\u003e\u003cp\u003e\u003cstrong\u003eA registered technical test or a manual release decision.\u003c\/strong\u003e It does not mean authority-approved, legally approved, or compliance with the law.\u003c\/p\u003e\n\u003ch3\u003eCan Excel or PDF create machine-readable marks?\u003c\/h3\u003e\u003cp\u003e\u003cstrong\u003eNo.\u003c\/strong\u003e Excel is a traceable control register. The PDF documents are searchable working documents; they do not add watermarks or metadata to the customer's own AI output.\u003c\/p\u003e\n\u003ch3\u003eHave the documents been reviewed by a lawyer?\u003c\/h3\u003e\u003cp\u003e\u003cstrong\u003eNo independent external legal review is warranted.\u003c\/strong\u003e AI-assisted source control of the relevant version and functional file tests are documented separately. For complex provider roles or exemptions, an individual assessment is required.\u003c\/p\u003e\n\u003ch2\u003eSources, dates, and transparency\u003c\/h2\u003e\n\u003cp\u003eVersion \u003cstrong\u003e1.0\u003c\/strong\u003e, document date \u003cstrong\u003e2026-11-09\u003c\/strong\u003e. AI-assisted source control 2026-11-09 against Regulation (EU) 2024\/1689 (consolidated 2026-07-27), especially Articles 50.2, 111.4, and 113, as well as the EU Commission guidelines for Article 50. The product is based on relevant public sources but is not authority-approved, certified, or provided with a guaranteed AI ranking.\u003c\/p\u003e","brand":"Svenska Dokumentmallar","offers":[{"title":"Default","offer_id":56563621822806,"sku":"AI50-2-PROVIDER-MARKING-V1","price":79.0,"currency_code":"SEK","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0827\/2373\/3846\/files\/ai50-2-maskinlasbar-markning-huvud-1600.png?v=1794265383","url":"https:\/\/mallbutiken.se\/en\/products\/ai-act-50-2-maskinlasbar-markning-detektion-testbevis","provider":"Mallbutiken","version":"1.0","type":"link"}