ukcopyrightartificial intelligenceintellectual propertygetty v stability

UK Copyright Policy Strategy & The Getty v Stability Appeal

Regulations.ai (AI-assisted)•

With the highly anticipated Getty v. Stability AI appeal arriving on October 1, 2026, artificial intelligence developers and content creators face a decisive moment in the United Kingdom. The government's Report on Copyright and Artificial Intelligence has made the official UK posture clear: there will be no statutory broad commercial text and data mining exception, leaving technology firms reliant on voluntary commercial licensing or narrow research exemptions.

What's changing

Published on March 18, 2026, under Section 137 of the Data (Use and Access) Act 2025, the UK government's policy report establishes a clear strategic direction for artificial intelligence and intellectual property. Rather than pushing for immediate, sweeping legislation or establishing a dedicated AI regulatory body, the government opted for a strategic pause on structural statutory overhauls. Instead, the strategy prioritizes voluntary market mechanisms, contractual licensing frameworks, and ongoing evidence gathering.

The policy report outlines several fundamental positions:

  • Abandonment of commercial TDM exceptions: The government formally dropped previous considerations to introduce a broad commercial text and data mining (TDM) exception paired with an opt-out mechanism. Commercial AI developers must continue obtaining explicit licences from rights holders to ingest copyrighted material.
  • Strict non-commercial TDM boundaries: Text and data mining exceptions under UK framework remain strictly limited to non-commercial scientific research.
  • Re-evaluating AI-generated content protection: Officials indicated a preference toward removing copyright protections currently granted under the Copyright, Designs and Patents Act 1988 (CDPA) for computer-generated works that lack a human author.
  • Protection against digital replicas: The government is examining potential future protections to counter unauthorized AI-generated imitations of an individual's voice, image, or likeness.
  • Market licensing infrastructure: Building on the launch of the Creative Content Exchange pilot on July 1, 2026, the strategy supports practical tools to help smaller creators and rights holders market and license digitized assets to AI developers.

Crucially, this policy report creates no new independent legal penalties or statutory enforcement bodies. Copyright compliance and enforcement continue under the existing Copyright, Designs and Patents Act 1988 through private civil litigation—including injunctions and financial damages—or criminal prosecution for commercial-scale infringement. Because the government has declined to create a commercial TDM safe harbor, upcoming court rulings like the Getty v. Stability AI appeal starting October 1, 2026, will serve as the primary legal tests shaping UK commercial training practices.

Who is affected

This policy applies at the national level across the United Kingdom. It impacts several core sectors:

  • AI model developers and technology platforms: Any startup, enterprise, or lab training foundation models or fine-tuning AI systems using UK-sourced data or operating commercially within the UK market.
  • Creative industries and rightsholders: Publishers, photographers, visual artists, record labels, and media platforms whose proprietary materials are contained in digital training corpuses.
  • Enterprise software buyers: Businesses acquiring or integrating synthetic media generation tools, who must manage downstream legal exposure and data provenance risks.

While smaller creators gain access to structured market channels through the Creative Content Exchange pilot, commercial technology companies face heightened exposure if they build commercial products using unlicenced UK content.

Three things to do this week

With the Getty v. Stability AI appeal set to open on October 1, 2026, corporate legal, product, and data governance teams should execute three immediate actions:

  1. Audit training data for commercial licence coverage: Conduct a comprehensive review of all datasets used for training, fine-tuning, or testing commercial models within the UK. Confirm that every dataset containing copyrighted material is backed by an explicit commercial licence rather than assumed fair dealing or an expected TDM exception.
  2. Isolate non-commercial research pipelines: Review TDM activities to ensure any rely on non-commercial research exemptions. Establish strict separation between experimental research environments and commercial product builds to avoid invalidating statutory non-commercial research protections.
  3. Assess IP ownership of pure synthetic outputs: Inventory internal media assets and product features that rely on fully computer-generated outputs. Prepare for prospective legal shifts that may eliminate copyright protection for works generated without meaningful human creative intervention.

Related context

This strategy document sits alongside several key UK framework milestones. It follows the groundwork laid out in the Copyright and artificial intelligence statement of progress under Section 137 Data (Use and Access) Act (/regulations/RAI-GB-NA-COPYRIG-2025) released in late 2025. It also operates within the context of the broader sectoral framework outlined in A pro-innovation approach to AI regulation: government response (/regulations/RAI-GB-NA-GRAWPXX-2024) and aligns with the evidentiary findings detailed in Data regulation in the age of AI and other data-intensive technologies (/regulations/RAI-GB-NA-EVIDENC-2026).

Note: this article was drafted by AI - Google Gemini

Back to News