ملاحظات (121) تا (140)

(121) Standardisation should play a key role to provide technical solutions to providers to ensure compliance with this Regulation, in line with the state of the art, to promote innovation as well as competitiveness and growth in the single market. Compliance with harmonised standards as defined in Article 2, point (1)(c), of Regulation (EU) No 1025/2012 of the European Parliament and of the Council (41), which are normally expected to reflect the state of the art, should be a means for providers to demonstrate conformity with the requirements of this Regulation. A balanced representation of interests involving all relevant stakeholders in the development of standards, in particular SMEs, consumer organisations and environmental and social stakeholders in accordance with Articles 5 and 6 of Regulation (EU) No 1025/2012 should therefore be encouraged. In order to facilitate compliance, the standardisation requests should be issued by the Commission without undue delay. When preparing the standardisation request, the Commission should consult the advisory forum and the Board in order to collect relevant expertise. However, in the absence of relevant references to harmonised standards, the Commission should be able to establish, via implementing acts, and after consultation of the advisory forum, common specifications for certain requirements under this Regulation. The common specification should be an exceptional fall back solution to facilitate the provider’s obligation to comply with the requirements of this Regulation, when the standardisation request has not been accepted by any of the European standardisation organisations, or when the relevant harmonised standards insufficiently address fundamental rights concerns, or when the harmonised standards do not comply with the request, or when there are delays in the adoption of an appropriate harmonised standard. Where such a delay in the adoption of a harmonised standard is due to the technical complexity of that standard, this should be considered by the Commission before contemplating the establishment of common specifications. When developing common specifications, the Commission is encouraged to cooperate with international partners and international standardisation bodies.

(122) It is appropriate that, without prejudice to the use of harmonised standards and common specifications, providers of a high-risk AI system that has been trained and tested on data reflecting the specific geographical, behavioural, contextual or functional setting within which the AI system is intended to be used, should be presumed to comply with the relevant measure provided for under the requirement on data governance set out in this Regulation. Without prejudice to the requirements related to robustness and accuracy set out in this Regulation, in accordance with Article 54(3) of Regulation (EU) 2019/881, high-risk AI systems that have been certified or for which a statement of conformity has been issued under a cybersecurity scheme pursuant to that Regulation and the references of which have been published in the Official Journal of the European Union should be presumed to comply with the cybersecurity requirement of this Regulation in so far as the cybersecurity certificate or statement of conformity or parts thereof cover the cybersecurity requirement of this Regulation. This remains without prejudice to the voluntary nature of that cybersecurity scheme.

(123) In order to ensure a high level of trustworthiness of high-risk AI systems, those systems should be subject to a conformity assessment prior to their placing on the market or putting into service.

(124) It is appropriate that, in order to minimise the burden on operators and avoid any possible duplication, for high-risk AI systems related to products which are covered by existing Union harmonisation legislation based on the New Legislative Framework, the compliance of those AI systems with the requirements of this Regulation should be assessed as part of the conformity assessment already provided for in that law. The applicability of the requirements of this Regulation should thus not affect the specific logic, methodology or general structure of conformity assessment under the relevant Union harmonisation legislation.

(125) Given the complexity of high-risk AI systems and the risks that are associated with them, it is important to develop an adequate conformity assessment procedure for high-risk AI systems involving notified bodies, so-called third party conformity assessment. However, given the current experience of professional pre-market certifiers in the field of product safety and the different nature of risks involved, it is appropriate to limit, at least in an initial phase of application of this Regulation, the scope of application of third-party conformity assessment for high-risk AI systems other than those related to products. Therefore, the conformity assessment of such systems should be carried out as a general rule by the provider under its own responsibility, with the only exception of AI systems intended to be used for biometrics.

(126) In order to carry out third-party conformity assessments when so required, notified bodies should be notified under this Regulation by the national competent authorities, provided that they comply with a set of requirements, in particular on independence, competence, absence of conflicts of interests and suitable cybersecurity requirements. Notification of those bodies should be sent by national competent authorities to the Commission and the other Member States by means of the electronic notification tool developed and managed by the Commission pursuant to Article R23 of Annex I to Decision No 768/2008/EC.

(127) In line with Union commitments under the World Trade Organization Agreement on Technical Barriers to Trade, it is adequate to facilitate the mutual recognition of conformity assessment results produced by competent conformity assessment bodies, independent of the territory in which they are established, provided that those conformity assessment bodies established under the law of a third country meet the applicable requirements of this Regulation and the Union has concluded an agreement to that extent. In this context, the Commission should actively explore possible international instruments for that purpose and in particular pursue the conclusion of mutual recognition agreements with third countries.

(128) In line with the commonly established notion of substantial modification for products regulated by Union harmonisation legislation, it is appropriate that whenever a change occurs which may affect the compliance of a high-risk AI system with this Regulation (e.g. change of operating system or software architecture), or when the intended purpose of the system changes, that AI system should be considered to be a new AI system which should undergo a new conformity assessment. However, changes occurring to the algorithm and the performance of AI systems which continue to ‘learn’ after being placed on the market or put into service, namely automatically adapting how functions are carried out, should not constitute a substantial modification, provided that those changes have been pre-determined by the provider and assessed at the moment of the conformity assessment.

(129) High-risk AI systems should bear the CE marking to indicate their conformity with this Regulation so that they can move freely within the internal market. For high-risk AI systems embedded in a product, a physical CE marking should be affixed, and may be complemented by a digital CE marking. For high-risk AI systems only provided digitally, a digital CE marking should be used. Member States should not create unjustified obstacles to the placing on the market or the putting into service of high-risk AI systems that comply with the requirements laid down in this Regulation and bear the CE marking.

(130) Under certain conditions, rapid availability of innovative technologies may be crucial for health and safety of persons, the protection of the environment and climate change and for society as a whole. It is thus appropriate that under exceptional reasons of public security or protection of life and health of natural persons, environmental protection and the protection of key industrial and infrastructural assets, market surveillance authorities could authorise the placing on the market or the putting into service of AI systems which have not undergone a conformity assessment. In duly justified situations, as provided for in this Regulation, law enforcement authorities or civil protection authorities may put a specific high-risk AI system into service without the authorisation of the market surveillance authority, provided that such authorisation is requested during or after the use without undue delay.

(131) In order to facilitate the work of the Commission and the Member States in the AI field as well as to increase the transparency towards the public, providers of high-risk AI systems other than those related to products falling within the scope of relevant existing Union harmonisation legislation, as well as providers who consider that an AI system listed in the high-risk use cases in an annex to this Regulation is not high-risk on the basis of a derogation, should be required to register themselves and information about their AI system in an EU database, to be established and managed by the Commission. Before using an AI system listed in the high-risk use cases in an annex to this Regulation, deployers of high-risk AI systems that are public authorities, agencies or bodies, should register themselves in such database and select the system that they envisage to use. Other deployers should be entitled to do so voluntarily. This section of the EU database should be publicly accessible, free of charge, the information should be easily navigable, understandable and machine-readable. The EU database should also be user-friendly, for example by providing search functionalities, including through keywords, allowing the general public to find relevant information to be submitted upon the registration of high-risk AI systems and on the use case of high-risk AI systems, set out in an annex to this Regulation, to which the high-risk AI systems correspond. Any substantial modification of high-risk AI systems should also be registered in the EU database. For high-risk AI systems in the area of law enforcement, migration, asylum and border control management, the registration obligations should be fulfilled in a secure non-public section of the EU database. Access to the secure non-public section should be strictly limited to the Commission as well as to market surveillance authorities with regard to their national section of that database. High-risk AI systems in the area of critical infrastructure should only be registered at national level. The Commission should be the controller of the EU database, in accordance with Regulation (EU) 2018/1725. In order to ensure the full functionality of the EU database, when deployed, the procedure for setting the database should include the development of functional specifications by the Commission and an independent audit report. The Commission should take into account cybersecurity risks when carrying out its tasks as data controller on the EU database. In order to maximise the availability and use of the EU database by the public, the EU database, including the information made available through it, should comply with requirements under the Directive (EU) 2019/882.

(132) Certain AI systems intended to interact with natural persons or to generate content may pose specific risks of impersonation or deception irrespective of whether they qualify as high-risk or not. In certain circumstances, the use of these systems should therefore be subject to specific transparency obligations without prejudice to the requirements and obligations for high-risk AI systems and subject to targeted exceptions to take into account the special need of law enforcement. In particular, natural persons should be notified 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. When implementing that obligation, the characteristics of natural persons belonging to vulnerable groups due to their age or disability should be taken into account to the extent the AI system is intended to interact with those groups as well. Moreover, natural persons should be notified when they are exposed to AI systems that, by processing their biometric data, can identify or infer the emotions or intentions of those persons or assign them to specific categories. Such specific categories can relate to aspects such as sex, age, hair colour, eye colour, tattoos, personal traits, ethnic origin, personal preferences and interests. Such information and notifications should be provided in accessible formats for persons with disabilities.

(133) A variety of AI systems can generate large quantities of synthetic content that becomes increasingly hard for humans to distinguish from human-generated and authentic content. The wide availability and increasing capabilities of those systems have a significant impact on the integrity and trust in the information ecosystem, raising new risks of misinformation and manipulation at scale, fraud, impersonation and consumer deception. In light of those impacts, the fast technological pace and the need for new methods and techniques to trace origin of information, it is appropriate to require providers of those systems to embed technical solutions that enable marking in a machine readable format and detection that the output has been generated or manipulated by an AI system and not a human. Such techniques and methods should be sufficiently reliable, interoperable, effective and robust as far as this is technically feasible, taking into account available techniques or a combination of such techniques, such as watermarks, metadata identifications, cryptographic methods for proving provenance and authenticity of content, logging methods, fingerprints or other techniques, as may be appropriate. When implementing this obligation, providers should also take into account the specificities and the limitations of the different types of content and the relevant technological and market developments in the field, as reflected in the generally acknowledged state of the art. Such techniques and methods can be implemented at the level of the AI system or at the level of the AI model, including general-purpose AI models generating content, thereby facilitating fulfilment of this obligation by the downstream provider of the AI system. To remain proportionate, it is appropriate to envisage that this marking obligation should not cover AI systems performing primarily an assistive function for standard editing or AI systems not substantially altering the input data provided by the deployer or the semantics thereof.

(134) Further to the technical solutions employed by the providers of the AI system, deployers who use an AI system to generate or manipulate image, audio or video content that appreciably resembles existing persons, objects, places, entities or events and would falsely appear to a person to be authentic or truthful (deep fakes), should also clearly and distinguishably disclose that the content has been artificially created or manipulated by labelling the AI output accordingly and disclosing its artificial origin. Compliance with this transparency obligation should not be interpreted as indicating that the use of the AI system or its output impedes the right to freedom of expression and the right to freedom of the arts and sciences guaranteed in the Charter, in particular where the content is part of an evidently creative, satirical, artistic, fictional or analogous work or programme, subject to appropriate safeguards for the rights and freedoms of third parties. In those cases, the transparency obligation for deep fakes set out in this Regulation is 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, including its normal exploitation and use, while maintaining the utility and quality of the work. In addition, it is also appropriate to envisage a similar disclosure obligation in relation to AI-generated or manipulated text to the extent it is published with the purpose of informing the public on matters of public interest unless the AI-generated content has undergone a process of human review or editorial control and a natural or legal person holds editorial responsibility for the publication of the content.

(135) Without prejudice to the mandatory nature and full applicability of the transparency obligations, the Commission may also 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, including to support practical arrangements for making, as appropriate, the detection mechanisms accessible and facilitating cooperation with other actors along the value chain, disseminating content or checking its authenticity and provenance to enable the public to effectively distinguish AI-generated content.

(136) The obligations placed on providers and deployers of certain AI systems in this Regulation to enable the detection and disclosure that the outputs of those systems are artificially generated or manipulated are particularly relevant to facilitate the effective implementation of Regulation (EU) 2022/2065. This applies in particular as regards the obligations of providers of very large online platforms or very large online search engines to identify and mitigate systemic risks that may arise from the dissemination of content that has been artificially generated or manipulated, in particular the risk of the actual or foreseeable negative effects on democratic processes, civic discourse and electoral processes, including through disinformation. The requirement to label content generated by AI systems under this Regulation is without prejudice to the obligation in Article 16(6) of Regulation (EU) 2022/2065 for providers of hosting services to process notices on illegal content received pursuant to Article 16(1) of that Regulation and should not influence the assessment and the decision on the illegality of the specific content. That assessment should be performed solely with reference to the rules governing the legality of the content.

(137) Compliance with the transparency obligations for the AI systems covered by this Regulation should not be interpreted as indicating that the use of the AI system or its output is lawful under this Regulation or other Union and Member State law and should be without prejudice to other transparency obligations for deployers of AI systems laid down in Union or national law.

(138) AI is a rapidly developing family of technologies that requires regulatory oversight and a safe and controlled space for experimentation, while ensuring responsible innovation and integration of appropriate safeguards and risk mitigation measures. To ensure a legal framework that promotes innovation, is future-proof and resilient to disruption, Member States should ensure that their national competent authorities establish at least one AI regulatory sandbox at national level to facilitate the development and testing of innovative AI systems under strict regulatory oversight before these systems are placed on the market or otherwise put into service. Member States could also fulfil this obligation through participating in already existing regulatory sandboxes or establishing jointly a sandbox with one or more Member States’ competent authorities, insofar as this participation provides equivalent level of national coverage for the participating Member States. AI regulatory sandboxes could be established in physical, digital or hybrid form and may accommodate physical as well as digital products. Establishing authorities should also ensure that the AI regulatory sandboxes have the adequate resources for their functioning, including financial and human resources.

(139) The objectives of the AI regulatory sandboxes should be to foster AI innovation by establishing a controlled experimentation and testing environment in the development and pre-marketing phase with a view to ensuring compliance of the innovative AI systems with this Regulation and other relevant Union and national law. Moreover, the AI regulatory sandboxes should aim to enhance legal certainty for innovators and the competent authorities’ oversight and understanding of the opportunities, emerging risks and the impacts of AI use, to facilitate regulatory learning for authorities and undertakings, including with a view to future adaptions of the legal framework, to support cooperation and the sharing of best practices with the authorities involved in the AI regulatory sandbox, and to accelerate access to markets, including by removing barriers for SMEs, including start-ups. AI regulatory sandboxes should be widely available throughout the Union, and particular attention should be given to their accessibility for SMEs, including start-ups. The participation in the AI regulatory sandbox should focus on issues that raise legal uncertainty for providers and prospective providers to innovate, experiment with AI in the Union and contribute to evidence-based regulatory learning. The supervision of the AI systems in the AI regulatory sandbox should therefore cover their development, training, testing and validation before the systems are placed on the market or put into service, as well as the notion and occurrence of substantial modification that may require a new conformity assessment procedure. Any significant risks identified during the development and testing of such AI systems should result in adequate mitigation and, failing that, in the suspension of the development and testing process. Where appropriate, national competent authorities establishing AI regulatory sandboxes should cooperate with other relevant authorities, including those supervising the protection of fundamental rights, and could allow for the involvement of other actors within the AI ecosystem such as national or European standardisation organisations, notified bodies, testing and experimentation facilities, research and experimentation labs, European Digital Innovation Hubs and relevant stakeholder and civil society organisations. To ensure uniform implementation across the Union and economies of scale, it is appropriate to establish common rules for the AI regulatory sandboxes’ implementation and a framework for cooperation between the relevant authorities involved in the supervision of the sandboxes. AI regulatory sandboxes established under this Regulation should be without prejudice to other law allowing for the establishment of other sandboxes aiming to ensure compliance with law other than this Regulation. Where appropriate, relevant competent authorities in charge of those other regulatory sandboxes should consider the benefits of using those sandboxes also for the purpose of ensuring compliance of AI systems with this Regulation. Upon agreement between the national competent authorities and the participants in the AI regulatory sandbox, testing in real world conditions may also be operated and supervised in the framework of the AI regulatory sandbox.

(140) This Regulation should provide the legal basis for the providers and prospective providers in the AI regulatory sandbox to use personal data collected for other purposes for developing certain AI systems in the public interest within the AI regulatory sandbox, only under specified conditions, in accordance with Article 6(4) and Article 9(2), point (g), of Regulation (EU) 2016/679, and Articles 5, 6 and 10 of Regulation (EU) 2018/1725, and without prejudice to Article 4(2) and Article 10 of Directive (EU) 2016/680. All other obligations of data controllers and rights of data subjects under Regulations (EU) 2016/679 and (EU) 2018/1725 and Directive (EU) 2016/680 remain applicable. In particular, this Regulation should not provide a legal basis in the meaning of Article 22(2), point (b) of Regulation (EU) 2016/679 and Article 24(2), point (b) of Regulation (EU) 2018/1725. Providers and prospective providers in the AI regulatory sandbox should ensure appropriate safeguards and cooperate with the competent authorities, including by following their guidance and acting expeditiously and in good faith to adequately mitigate any identified significant risks to safety, health, and fundamental rights that may arise during the development, testing and experimentation in that sandbox.