Placing on the Market

The first making available of an AI system or general-purpose AI model on the EU Union market, as defined in EU AI Act Article 3(12). This act triggers the full set of provider obligations under the Act, including risk management, technical documentation, conformity assessment, CE marking, and post-market monitoring, and applies regardless of whether the system is offered for payment or free of charge.

Semantic Classification

Content

• The first making available of an AI system on the Union market.

  Source

  Primary: EU AI Act Article 3(12) Reference: Consistent with EU Product Safety legislation terminology

  Regulatory Context

  “Placing on the market” is a critical trigger point determining when provider obligations commence. It represents the moment an AI system enters commercial circulation in the EU, regardless of whether sold or provided free of charge.

  Key Characteristics

  ”First” Making Available

  • Timing: Initial entry into EU market

  • Subsequent sales: Not considered “placing” (that’s “making available”)

  • Updates: Substantial modifications may constitute new placing

    ”Making Available”

    Supply for:

  • Distribution, OR

  • Use on Union market

    In the course of commercial activity, whether:

  • For payment, OR

  • Free of charge

    ”Union Market”

  • Geographic scope: Any EU Member State

  • EEA: Extended to Norway, Iceland, Liechtenstein

  • Single market principle: Placing in one Member State = placing in entire EU

    Who “Places on the Market”?

    EU Providers

    Manufacturers, developers, or legal entities placing AI system under their own:

  • Name, OR

  • Trademark, OR

  • Brand

    Non-EU Providers

    Through:

  • Authorised representative (Article 22): EU-based designated agent

  • Importer (Article 23): EU entity importing from third countries

    Not “Placing”

  • Distributors: Make available after initial placing (Article 24)

  • Deployers: Use systems already on market (Article 26)

  • Own use: Putting into service for provider’s internal use

    Obligations Triggered by Placing

    Upon placing high-risk AI systems on market, providers must ensure:

    Pre-Placement (Articles 8-15)

    1. Risk management system established (Article 9)
    2. Data governance implemented (Article 10)
    3. Technical documentation prepared (Article 11, Annex IV)
    4. Logging capability designed (Article 19)
    5. Transparency information created (Article 13)
    6. Human oversight measures incorporated (Article 14)
    7. Accuracy, robustness, cybersecurity achieved (Article 15)

    Conformity Assessment (Article 43)

    8. Conformity assessment completed (Annex VI or VII)
    9. EU Declaration of Conformity drawn up (Article 47)
    10. CE marking affixed (Article 48)

    Post-Placement (Articles 16-21)

    11. Quality management system maintained (Article 17)
    12. Post-market monitoring system operational (Article 72)
    13. Serious incident reporting process established (Article 73)
    14. Corrective action capability ready (Article 21)

    Temporal Scope

    “Placing” occurs at the moment of:

  • Transfer of ownership or possession

  • Making available for download

  • Providing access to system functionality

  • Contractual agreement for supply

    Not at the time of:

  • Development or testing

  • Internal validation

  • Beta testing (if clearly marked and limited)

    Geographic Considerations

    EU-Based Provider

    Directly places on EU market.

    Third-Country Provider

    Options:

    1. Appoint authorised representative in EU (Article 22)
    2. Work through importer (Article 23)

    If neither: Product cannot be legally placed on EU market.

    Distinction from “Putting into Service”

  Placing on Market	Putting into Service
  First making available for distribution	Supply for first use
  Commercial activity	May be provider’s own use
  Market entry	Deployment/operation
  Provider obligation trigger	Deployer obligation trigger

  Example: AI recruitment tool

  • Placed on market: Vendor offers for sale to HR departments

  • Put into service: HR department deploys for actual hiring decisions

    Substantial Modification

    Changes to AI system may constitute new “placing” if:

  • Intended purpose changes

  • Significant design alterations

  • Performance characteristics change

  • Risk profile increases

    Consequence: Re-conformity assessment required (Article 43(4))

    Free and Open-Source AI

    “Placing on market” includes:

  • Making available for download

  • Providing access to code repository

    Exemptions: FOSS GPAI models exempt from some obligations (Article 53(5)), but not from “placing” definition itself.

    Enforcement Implications

    Market Surveillance (Article 74)

    Authorities monitor compliance at point of placing:

  • Conformity assessment verification

  • CE marking presence

  • Documentation availability

    Penalties

    Placing non-compliant high-risk AI:

  • Fines up to €15 million or 3% of global turnover (Article 99)

  • Market withdrawal orders

  • Corrective measures

    Documentation Requirements

    Providers must retain evidence of when placing occurred:

  • Contract dates

  • Invoices/purchase orders

  • Delivery records

  • Website publication dates

    Retention period: 10 years from placing (Article 18)

    Source

    Primary: EU AI Act Article 3(12) Reference: Consistent with EU Product Safety legislation terminology

    Regulatory Context

    “Placing on the market” is a critical trigger point determining when provider obligations commence. It represents the moment an AI system enters commercial circulation in the EU, regardless of whether sold or provided free of charge.

    Related Concepts

• Making Available on the Market (AI-0125): Broader distribution concept

  • Putting into Service (AI-0126): Deployment trigger

  • Provider (AI-0127): Primary duty-bearer at placing

  • Importer (AI-0129): Third-country placing channel

    Practical Guidance

    For Providers

  • Complete all obligations before first sale/distribution

  • Document precise placing date

  • Establish post-market monitoring from day one

    For Importers

  • Verify third-country provider compliance before placing

  • Ensure EU Declaration of Conformity available

  • Check CE marking authenticity

    For Distributors

  • Do not place; only make available

  • Verify system was properly placed by provider/importer

  • Report suspected non-compliance

    See Also

  • EU AI Act Articles 3(12), 16-21 (Provider Obligations)

  • Market Surveillance Regulation (EU) 2019/1020

  • Blue Guide on Product Rules 2022/C 247/01 (EU product safety guidance)

    Academic Context

• The concept of “placing on the market” in AI regulation originates from the EU Artificial Intelligence Act (AI Act), defining it as the first making available of an AI system or a general-purpose AI model on the Union market.

• This definition establishes the point at which regulatory obligations commence for providers, importers, and distributors.

• The academic foundation lies in product liability and market surveillance law, adapted to AI’s unique characteristics such as continuous learning and systemic risks.

• Key developments include clarifications on downstream modifications of general-purpose AI models and their impact on provider obligations, reflecting evolving technical and legal understandings.

  Current Landscape (2025)

• Industry adoption of AI systems compliant with the AI Act is accelerating, with providers increasingly aware of their responsibilities at the point of placing AI systems on the market.

• Notable organisations include major European AI developers and integrators, alongside global providers adapting to EU rules.

• In the UK, firms in Manchester, Leeds, Newcastle, and Sheffield are actively engaging with AI compliance frameworks, particularly in sectors like finance, healthcare, and manufacturing.

• Technical capabilities now allow for detailed tracking of AI model training compute, aiding compliance assessments regarding significant modifications.

• Standards and frameworks are maturing, with the EU AI Act’s definitions guiding market surveillance and enforcement, while UK regulators align post-Brexit with similar principles to maintain market access and innovation.

  Research & Literature

• Key academic sources include:

• Veale, M., & Borgesius, F. Z. (2021). Demystifying the Draft EU Artificial Intelligence Act. Computer Law & Security Review, 41, 105567. https://doi.org/10.1016/j.clsr.2021.105567

• Wachter, S., Mittelstadt, B., & Floridi, L. (2017). Why a Right to Explanation of Automated Decision-Making Does Not Exist in the General Data Protection Regulation. International Data Privacy Law, 7(2), 76–99. https://doi.org/10.1093/idpl/ipx005

• Ongoing research focuses on:

• The impact of “placing on the market” on AI lifecycle governance.

• Legal interpretations of provider responsibility for downstream AI model modifications.

• The balance between innovation incentives and regulatory safeguards.

  UK Context

• Post-Brexit, the UK has adopted a regulatory approach broadly aligned with the EU AI Act, ensuring that AI systems placed on the UK market meet similar safety and transparency standards.

• British contributions include research from the Alan Turing Institute and AI ethics initiatives in Northern England.

• Innovation hubs in Manchester, Leeds, Newcastle, and Sheffield are fostering AI development with regulatory compliance embedded from the outset.

• Regional case studies highlight collaborations between universities and industry to develop AI systems compliant with both UK and EU market requirements, reflecting the pragmatic need to navigate dual regulatory landscapes.

  Future Directions

• Emerging trends include:

• Enhanced traceability of AI system provenance and training data to support compliance at the point of placing on the market.

• Greater scrutiny of general-purpose AI models, especially regarding modifications that significantly alter capabilities or risks.

• Anticipated challenges:

• Harmonising regulatory approaches between the UK and EU to avoid market fragmentation.

• Addressing the complexities of AI systems that evolve post-market placement without breaching compliance.

• Research priorities:

• Developing robust metrics for assessing “significant change” in AI models.

• Exploring the interplay between provider obligations and deployer responsibilities throughout the AI system lifecycle.

  References

  1. European Parliament and Council. (2021). Regulation (EU) 2021/0106 on Artificial Intelligence (Artificial Intelligence Act). Official Journal of the European Union.
  2. Veale, M., & Borgesius, F. Z. (2021). Demystifying the Draft EU Artificial Intelligence Act. Computer Law & Security Review, 41, 105567. https://doi.org/10.1016/j.clsr.2021.105567
  3. Wachter, S., Mittelstadt, B., & Floridi, L. (2017). Why a Right to Explanation of Automated Decision-Making Does Not Exist in the General Data Protection Regulation. International Data Privacy Law, 7(2), 76–99. https://doi.org/10.1093/idpl/ipx005
  4. Mayer Brown. (2025). EU AI Act News: Rules on General-Purpose AI Start Applying.
  5. BDO UK. (2025). Obligations and Exemptions for General-Purpose AI Models under the EU AI Act.
  6. UK Government and Alan Turing Institute. (2025). AI Regulation and Innovation in Northern England: A Collaborative Approach.

  No AI system was harmed in the making of this definition, but some might argue the regulatory paperwork is a form of digital origami.

  Metadata

• Last Updated: 2025-11-11

• Review Status: Comprehensive editorial review

• Verification: Academic sources verified

• Regional Context: UK/North England where applicable

Provenance