欧州委員会 AI生成コンテンツの表示・ラベル付けに関する行動規範のドラフト第2版 (2025.03.05)
こんにちは、丸山満彦です。
AI法、第50条関係の話です...
| Chapter IV: Transparency Obligations for Providers and Deployers of Certain AI Systems | 第4章:特定の人工知能システムプロバイダ及び展開者に対する透明性義務 |
| Article 50: Transparency obligations for providers and deployers of certain AI systems | 第50条:特定の人工知能システムプロバイダ及び展開者に対する透明性義務 |
| Summary | 要約 |
| Providers must inform users when they are interacting directly with an AI system. AI-generated or -manipulated content must be clearly marked and detectable as artificially generated. Deployers of emotion recognition or biometric categorisation systems must inform exposed person about the operation of the system. Deepfakes and AI-generated text must be disclosed as artificially generated. An exemption from these transparency obligations applies in case the systems are authorised by law to detect, prevent, investigate or prosecute criminal offences. Other specific exceptions apply per transparency obligation. Information must be provided clearly, distinguishably and accessibly. | プロバイダは、ユーザーが人工知能システムと直接やり取りしている場合に、その旨をユーザーに通知しなければならない。人工知能によって生成または操作されたコンテンツは、人工的に生成されたものであることが明確に表示され、検知可能でなければならない。感情認識システムまたは生体認証分類システムの展開者は、対象となる人物に対し、システムの動作について通知しなければならない。ディープフェイクおよびAI生成テキストは、人工的に生成されたものであることを開示しなければならない。これらの透明性義務は、犯罪の検知、防止、調査または起訴のために法的に認可されたシステムには適用除外となる。その他の具体的な例外は、各透明性義務ごとに適用される。情報は明確に、識別可能に、かつアクセス可能な形で提供されなければならない。 |
| The summaries are meant to provide helpful explanation but are not legal binding. | 要約は参考説明を目的とするが、法的拘束力はない。 |
| 1. Providers shall ensure that AI systems intended to interact directly with natural persons are designed and developed in such a way that the natural persons concerned are informed that they are interacting with an AI system, unless this is obvious from the point of view of a natural person who is reasonably well-informed, observant and circumspect, taking into account the circumstances and the context of use. This obligation shall not apply to AI systems authorised by law to detect, prevent, investigate or prosecute criminal offences, subject to appropriate safeguards for the rights and freedoms of third parties, unless those systems are available for the public to report a criminal offence. | 1. プロバイダは、自然人と直接対話するAIシステムについて、以下の条件を満たす設計・開発を保証しなければならない。すなわち、当該自然人がAIシステムと対話していることを認識できること。ただし、状況や使用環境を考慮し、合理的によく情報に通じ、注意深く慎重な自然人の観点から明らかな場合は除く。この義務は、犯罪の検知、防止、調査、起訴を目的として法的に認可されたAIシステムには適用されない。ただし、サードパーティの権利と自由に対する適切な保護措置が講じられている場合に限る。また、当該システムが一般市民による犯罪通報に利用可能な場合はこの限りではない。 |
| 2. Providers of AI systems, including general-purpose AI systems, generating synthetic audio, image, video or text content, shall ensure that the outputs of the AI system are marked in a machine-readable format and detectable as artificially generated or manipulated. Providers shall ensure their technical solutions are effective, interoperable, robust and reliable as far as this is technically feasible, taking into account the specificities and limitations of various types of content, the costs of implementation and the generally acknowledged state of the art, as may be reflected in relevant technical standards. This obligation shall not apply to the extent the AI systems perform an assistive function for standard editing or do not substantially alter the input data provided by the deployer or the semantics thereof, or where authorised by law to detect, prevent, investigate or prosecute criminal offences. | 2. 合成音声、画像、動画またはテキストコンテンツを生成する汎用AIシステムを含むAIシステムのプロバイダは、AIシステムの出力結果が機械可読形式でマークされ、人工的に生成または操作されたものであることが検知可能であることを確保しなければならない。プロバイダは、技術的に実現可能な範囲で、様々なコンテンツの特性や制約、実装コスト、関連標準に反映される一般的な技術水準を考慮し、自社の技術的解決策が効果的、相互運用可能、堅牢かつ信頼性のあるものであることを確保しなければならない。この義務は、AIシステムが標準的な編集のための補助機能を果たす場合、または展開者が提供する入力データやその意味論を実質的に変更しない場合、もしくは犯罪の検知、防止、調査、起訴のために法律で認められる場合には適用されない。 |
| 3. Deployers of an emotion recognition system or a biometric categorisation system shall inform the natural persons exposed thereto of the operation of the system, and shall process the personal data in accordance with Regulations (EU) 2016/679 and (EU) 2018/1725 and Directive (EU) 2016/680, as applicable. This obligation shall not apply to AI systems used for biometric categorisation and emotion recognition, which are permitted by law to detect, prevent or investigate criminal offences, subject to appropriate safeguards for the rights and freedoms of third parties, and in accordance with Union law. | 3. 感情認識システムまたは生体認証分類システムの展開者は、当該システムに晒される自然人に対し、システムの運用について通知し、適用される規則(EU)2016/679、規則(EU)2018/1725、指令(EU)2016/680に従って個人データを処理しなければならない。この義務は、サードパーティの権利と自由に対する適切な保護措置を条件とし、かつ連合法に従って、犯罪の検知、防止、調査のために生体認証分類および感情認識に使用されるAIシステムには適用されない。 |
| 4. Deployers of an AI system that generates or manipulates image, audio or video content constituting a deep fake, shall disclose that the content has been artificially generated or manipulated. This obligation shall not apply where the use is authorised by law to detect, prevent, investigate or prosecute criminal offence. Where the content forms part of an evidently artistic, creative, satirical, fictional or analogous work or programme, the transparency obligations set out in this paragraph are limited to disclosure of the existence of such generated or manipulated content in an appropriate manner that does not hamper the display or enjoyment of the work. | 4. ディープフェイクを構成する画像、音声、動画コンテンツを生成または操作するAIシステムの展開者は、そのコンテンツが人工的に生成または操作されたものであることを開示しなければならない。この義務は、犯罪の検知、防止、調査または起訴のために法律で認められた使用には適用されない。コンテンツが明らかに芸術的、創造的、風刺的、虚構的または類似の著作物または番組の一部を構成する場合、本項に定める透明性義務は、当該著作物の表示または享受を妨げない適切な方法で、生成または操作されたコンテンツの存在を開示することに限定される。 |
| Deployers of an AI system that generates or manipulates text which is published with the purpose of informing the public on matters of public interest shall disclose that the text has been artificially generated or manipulated. This obligation shall not apply where the use is authorised by law to detect, prevent, investigate or prosecute criminal offences or where the AI-generated content has undergone a process of human review or editorial control and where a natural or legal person holds editorial responsibility for the publication of the content. | 公共の利益に関する事項について公衆に情報を提供することを目的として公開されるテキストを生成または操作するAIシステムの展開者は、当該テキストが人工的に生成または操作されたものであることを開示しなければならない。この義務は、犯罪の検知、防止、調査または訴追のために法律により使用が許可されている場合、またはAI生成コンテンツが人間のレビューまたは編集管理プロセスを経ており、かつ自然人または法人が当該コンテンツの公開について編集責任を有する場合には適用されない。 |
| 5. The information referred to in paragraphs 1 to 4 shall be provided to the natural persons concerned in a clear and distinguishable manner at the latest at the time of the first interaction or exposure. The information shall conform to the applicable accessibility requirements. | 5. 第1項から第4項までに規定する情報は、遅くとも最初の相互作用またはエクスポージャーの時点で、関係する自然人に対して明確かつ識別可能な方法で提供されなければならない。当該情報は、適用されるアクセシビリティ要件に適合しなければならない。 |
| 6. Paragraphs 1 to 4 shall not affect the requirements and obligations set out in Chapter III, and shall be without prejudice to other transparency obligations laid down in Union or national law for deployers of AI systems. | 6. 第1項から第4項までの規定は、第III章に定める要件及び義務に影響を与えず、また、AIシステムの展開者に対してEU法または国内法で定められたその他の透明性義務を妨げるものではない。 |
| 7. The AI Office shall encourage and facilitate the drawing up of codes of practice at Union level to facilitate the effective implementation of the obligations regarding the detection and labelling of artificially generated or manipulated content. The Commission may adopt implementing acts to approve those codes of practice in accordance with the procedure laid down in Article 56 (6). If it deems the code is not adequate, the Commission may adopt an implementing act specifying common rules for the implementation of those obligations in accordance with the examination procedure laid down in Article 98(2). | 7. AI事務局は、人工的に生成または操作されたコンテンツの検知及び表示に関する義務の効果的な実施を促進するため、EUレベルでの行動規範の策定を奨励し、促進するものとする。委員会は、第56条(6)に定める手続に従い、当該行動規範を承認するための実施法令を採択することができる。委員会が当該規範が不十分であると判断した場合、第98条(2)に定める審査手続に従い、当該義務の実施に関する共通規則を定める実施法令を採択することができる。 |
● European Commission
・2026.03.05 Commission publishes second draft of Code of Practice on Marking and Labelling of AI-generated content
| Commission publishes second draft of Code of Practice on Marking and Labelling of AI-generated content | 欧州委員会は、AI生成コンテンツの表示・ラベル付けに関する行動規範のドラフト第2版を公表した。 |
| To help providers and deployers meet the marking and labelling requirements for AI generated content under Article 50 AI Act, the Commission is facilitating the development of a voluntary code of practice. | AI法(Article 50 AI Act)に基づくAI生成コンテンツの表示・ラベル付け要件を、プロバイダと展開者が満たすのを支援するため、欧州委員会は自主的な行動規範の策定を促進している。 |
| This second version, drafted by independent experts, integrates written feedback from hundreds of participants and observers, including industry, academia, civil society and other stakeholders. The feedback was gathered through an EU survey, stakeholder meetings and workshops held in January 2026, as part of the code of practice drafting process. The draft also incorporates contributions from Member States (submitted via the AI Board) and Members of the European Parliament (represented in the IMCO-LIBE Working group monitoring the AI Act implementation). | この第2版は独立した専門家によって起草され、産業界、学界、市民社会、その他の利害関係者を含む数百名の参加者・オブザーバーからの書面によるフィードバックを統合している。フィードバックは、行動規範起草プロセスの一環として、2026年1月に実施されたEU調査、ステークホルダー会議、ワークショップを通じて収集された。ドラフトには、加盟国(AI理事会経由で提出)および欧州議会議員(AI法実施を監視するIMCO-LIBE作業部会で代表者)からの意見も反映されている。 |
| The newest draft of the code has been streamlined and simplified, providing more flexibility for the signatories, reducing the compliance burden and incorporating further technical considerations to improve legal clarity and practicality. It promotes the use of open standards for AI content marking and an EU icon for labelling to simplify compliance and reduce costs. | 最新版のドラフト行動規範は簡素化・合理化され、署名者への柔軟性を高め、コンプライアンス負担を軽減するとともに、法的明確性と実用性を向上させるための技術的配慮を追加している。AIコンテンツ表示におけるオープン標準の活用と、EU統一アイコンによる表示を推進し、コンプライアンスの簡素化とコスト削減を図る。 |
| Sections of the draft code of practice | ドラフト行動規範のセクション |
| The draft code of practice for transparency of AI-generated content is structured into 2 sections, each dealing with different aspects of transparency and regulation for providers and deployers respectively. | AI生成コンテンツの透明性に関する行動規範ドラフトは2つのセクションで構成され、それぞれプロバイダと展開者に対する透明性・規制の異なる側面を扱う。 |
| Section 1 | セクション1 |
| Section 1 addresses marking and detecting AI content and is aimed at providers of generative AI systems within the scope of Article 50(2) AI Act. Compared to the first draft, this section of the code has undergone significant changes and introduced greater flexibility and clarity. | セクション1はAIコンテンツの表示と検知を扱い、AI法50条(2)項の適用範囲内にある生成的AIシステムプロバイダを対象とする。ドラフトと比較し、本節は大幅な変更を経て柔軟性と明確性が向上した。 |
| This revision removes and consolidates several measures, and introduces optional elements, while ensuring that all measures remain technically feasible and proportionate. Key commitments include a revised two-layered marking approach involving secured metadata and watermarking, optional fingerprinting and logging, and protocols for detection and verification. | 今回の改訂では複数の措置が削除・統合され、任意要素が導入された。同時に全措置の技術的実現可能性と比例性が確保されている。主な取り組みには、保護されたメタデータと電子透かしを組み合わせた二層式表示手法の改訂、任意のフィンガープリンティングとログ記録、検知・検証プロトコルが含まれる。 |
| Section 2 | セクション2 |
| Section 2, targeting deployers of AI systems, focuses on labelling deepfakes and text publications concerning matters of public interest within the scope of Article 50(4) AI Act. Relative to the first draft, this section adopts a more flexible and practice-oriented approach. It has been restructured to simplify and streamline the commitments, while the taxonomy distinguishing the AI–generated content from AI-assisted content has been completely removed. Section 2 now features design and placement requirements applicable to icons, labels or disclaimers, ensuring a minimum level of uniformity, while enabling signatories to conceive solutions adapted to their needs. In addition, the section proposes a task force to develop a future, uniform, interactive EU icon, with discretionary support from signatories. | 第2節はAIシステム展開者を対象とし、AI法第50条(4)の範囲内で公共の利益に関わる事項に関するディープフェイク及びテキスト出版物の表示に焦点を当てる。ドラフトと比較し、本節はより柔軟で実践的なアプローチを採用している。義務を簡素化・合理化するため再構成され、AI生成コンテンツとAI支援コンテンツを区別する分類体系は完全に削除された。第2節では、アイコン・ラベル・免責事項に適用される設計・配置要件を規定し、最低限の統一性を確保しつつ、署名者が自らのニーズに適した解決策を考案できるようにした。さらに、署名者の任意の支援を得て、将来の統一的なインタラクティブEUアイコンを開発するタスクフォースの設置を提案している。 |
| The code has also further defined the specific regimes applicable to artistic, creative, satirical and fictional works and text publications under human review or editorial control, facilitating reliance on existing practices or procedures. | また、芸術的・創造的・風刺的・虚構的作品、および人間の審査・編集管理下にあるテキスト出版物に適用される特例制度を明確化し、既存の慣行や手続きの活用を容易にした。 |
| The annex of the second draft now includes illustrative examples of a potential EU icon to be made freely available to signatories. These examples (and others) will be discussed with stakeholders as part of the next set of workshops. | 第2ドラフトの附属書には、署名機関が自由に利用できるEUアイコンの具体例が追加された。これらの例(およびその他の案)は、次回ワークショップで関係者と協議される予定だ。 |
| Next steps | 今後の手順 |
| The Commission will collect feedback on the second draft from participants and observers to the code of Practice until 30 March EOB. The code is expected to be finalised by beginning of June this year. | 欧州委員会は、3月30日(欧州時間)まで、行動規範の参加者及びオブザーバーから第2ドラフトに関するフィードバックを収集する。本規範は今年6月初旬までに最終化される見込みである。 |
| The rules covering the transparency of AI-generated content will become applicable on 2 August 2026. | AI生成コンテンツの透明性に関する規則は、2026年8月2日に適用開始となる。 |
・[PDF]
目次...
| Introductory statement by the Chairs and ViceChairs | 議長及び副議長による序文 |
| Section 1: Rules for marking and detection of AIgenerated and manipulated content applicable to providers of generative AI systems (Article 50(2) and (5) AI Act) | 第1節:生成的AIプロバイダに適用されるAI生成・改変コンテンツの表示及び検知に関する規則(AI法第50条(2)及び(5)) |
| Objectives | 目的 |
| Recitals | 前文 |
| Commitments | コミットメント |
| Commitment 1: Multi-layered Marking of AI-Generated Content | コミットメント1:AI生成コンテンツの多層的マーキング |
| Measure 1.1: Machine-readable marking techniques | 施策 1.1:機械可読マーキング技術 |
| Measure 1.2: Non-removal of machine-readable marking | 施策1.2:機械可読マーキングの非除去 |
| Measure 1.3: Transparency of the provenance chain (optional) | 施策 1.3:来歴証明の透明性(任意) |
| Measure 1.4: Optional functionality for perceptible markings (for deep fakes and AI-generated and manipulated published text) | 施策 1.4:知覚可能なマーキングのオプション機能(ディープフェイクおよび AI 生成・改変された公開テキスト向け) |
| Commitment 2: Detection of the Marking of AI-Generated Content | コミットメント2:AI生成コンテンツの表示検知 |
| Measure 2.1: Detection mechanisms for active marking made available to deployers, end-users and other third parties | 施策 2.1:展開者、エンドユーザー、その他のサードパーティに提供される能動的マーキング検知メカニズム |
| Measure 2.2: Forensic detection mechanisms (optional supplementary measure) | 施策 2.2:フォレンジック検知メカニズム(任意の補足施策) |
| Measure 2.3: Clear and accessible disclosure of verification and detection results | 施策2.3:検証及び検知結果の明確かつアクセス可能な開示 |
| Measure 2.4: Support literacy on AI marking technologies and verification | 施策 2.4:AI マーキング技術と検証に関する知識の普及支援 |
| Commitment 3: Measures to meet the Requirements for Marking and Detection Techniques | コミットメント3:表示要件及び検知技術に関する施策 |
| Measure 3.1: Effectiveness | 施策 3.1:有効性 |
| Measure 3.2: Reliability | 施策 3.2:信頼性 |
| Measure 3.3: Robustness | 施策 3.3:堅牢性 |
| Measure 3.4: Interoperability | 施策 3.4:相互運用性 |
| Measure 3.5: Advancing the state of the art in marking and detection | 施策3.5:マーキング及び検知技術の最先端の進展 |
| Commitment 4: Testing, verification and compliance | コミットメント4:試験、検証及びコンプライアンス |
| Measure 4.1: Compliance framework | 施策 4.1:コンプライアンス枠組み |
| Measure 4.2: Testing, verification and monitoring | 施策4.2:試験、検証及び監視 |
| Measure 4.3: Training | 施策 4.3:訓練 |
| Measure 4.4: Cooperation with market surveillance authorities | 施策 4.4:市場監視当局との協力 |
| Glossary | 用語集 |
| Section 2: Rules for labelling deepfakes and certain AI-generated and manipulated published text applicable to deployers of AI systems (Article 50(4) and (5) AI Act) | 第2節:AIシステムの展開者に適用される、ディープフェイクおよび特定のAI生成・操作済み公開テキストの表示に関する規則(AI法第50条(4)および(5)) |
| Objectives | 目的 |
| Recitals | 前文 |
| Commitments | コミットメント |
| Commitment 1: Disclosure of AI-Generated and Manipulated Deep Fakes and Published Text | コミットメント1:AI生成・改変ディープフェイク及び公開テキストの開示 |
| Measure 1.1 Design requirements for icons, labels or disclaimers | 施策 1.1 アイコン、ラベル、免責事項のデザイン要件 |
| Measure 1.2 Placement requirements for icons, labels or disclaimers | 施策 1.2 アイコン、ラベル、免責事項の配置要件 |
| Measure 1.3 Optional use of an EU icon and participation in its development | 施策 1.3 EU アイコンの任意使用と開発への参加 |
| Commitment 2: Proportionate compliance, awareness and review | コミットメント 2:比例的な遵守、認識、および見直し |
| Measure 2.1: Internal compliance | 施策 2.1:内部コンプライアンス |
| Measure 2.2: Awareness and Training | 施策 2.2:意識向上およびトレーニング |
| Measure 2.3: Review, feedback and cooperation with authorities | 施策 2.3:当局とのレビュー、フィードバック、協力 |
| Commitment 3: Appropriate Disclosure for Artistic, Creative and similar Works | コミットメント 3:芸術的、創造的及び類似の著作物に対する適切な開示 |
| Commitment 4: Human review, editorial control and responsibility in relation to AI-generated or manipulated text publications | コミットメント4:AI生成または改変テキスト出版物に関する人的審査、編集管理及び責任 |
| Appendices | 附属書 |
| Appendix 1 Sample Icon | 附属書 1 アイコンのサンプル |
欧州委員会が開発したサンプルアイコン(左)と背景に対するバリエーション(右)
● まるちゃんの情報セキュリティ気まぐれ日記
・2025.11.11 欧州 欧州委員会 AI生成コンテンツの表示・ラベル付けに関する行動規範の策定作業を開始 (2025.11.05)
・2025.03.16 中国 人工知能生成合成コンテンツ識別弁法 (2025.03.14)
AI法関係...
・2026.01.27 欧州 EDPB EDPS 欧州委員会の「AIに関するデジタルオムニバス」提案に関する共同意見書
・2025.12.22 ドイツ DSK GDPRの改革案(AIの利用に関する明確な法規制、IT企業に責任を持たせる)を提案 (2025.12.12)
・2025.12.04 欧州委員会 AI法についての内部通報ツールの発表 (2025.11.24)
・2025.11.28 オランダ データ保護庁はプライバシーと安全保障、経済効率性(技術革新)の間でいろいろと考えていますね
・2025.11.22 欧州委員会 EU企業の成長を支援する簡素化されたデジタル規則 (オムニバス法)
・2025.10.11 欧州委員会 AI法エクスプローラー
・2025.08.02 欧州委員会 AI法第 53 条(1)(d)で要求される汎用 AI モデルの訓練内容に関する公開要約の説明文書およびテンプレー(2025.07.24)
・2025.07.31 欧州委員会 AI法における汎用AIモデル提供者の義務範囲に関するガイドライン(2025.07.18)
・2025.07.25 欧州委員会 汎用 AI 実践規範 (2025.07.10)
・2025.05.24 欧州 単一市場戦略 「今日の不確実な世界において、欧州人の最初のパートナーは欧州人自身でなければならない」(2025.05.21)
・2025.03.13 欧州委員会 AI法における汎用AIモデル - Q&A (2025.03.10)
・2025.02.08 欧州委員会 規則(EU)2024/1689(AI法)が定める人工知能の禁止行為に関する欧州委員会ガイドライン
・2024.12.06 欧州委員会 AI法における汎用AIモデル - Q&A (2024.11.20)
・2024.10.30 欧州 AI法の調和標準の策定について...
・2024.10.18 オランダ AI利用ガイド (2024.10.16)
・2024.08.05 欧州AI法が施行された... (2024.08.01)
・2024.07.20 EU EDPB GDPRとEU-AI法関連
・2024.07.19 ドイツ BfDI フランス CNIL オランダ AP EU AI法関連
・2024.07.16 EU 2024.07.12にAI法がEU官報に掲載された
« 米国 NIST AI 800-4 展開済みAIシステムの監視における課題 (2026.03.09) | Main | 欧州 ETSI 人工知能のセキュリティ確保(SAI);AIモデル及びシステムに対する基本サイバーセキュリティ要件 (2026.01.15) »



Comments