16 April 2026

Copyright & artificial intelligence: Progress, pause and persistent uncertainty

The UK Government has now published its March 2026 Report on Copyright and Artificial Intelligence, following its 2024–25 consultation on the use of copyright‑protected works in AI training. The outcome is significant – not for what it introduces but for what it postpones.

For now, copyright reform has been paused. The question is whether this pause meaningfully improves certainty for creators and AI developers or simply prolongs an already difficult status quo.

Why this consultation mattered

The consultation addressed a genuine and growing problem. Rights holders remain concerned that their works are being used to train AI systems without consent or remuneration. AI developers, meanwhile, stress that access to large volumes of data is essential for innovation and competitiveness.

The Government accepted that UK copyright law, drafted long before generative AI, does not provide clear answers. In practice, businesses have been left to navigate this through litigation, licensing workarounds and varying risk appetites.

That uncertainty carries real economic and operational consequences.

The Government’s original objectives

Launched in December 2024, the consultation set out three aims:

  1. To support rights holders’ ability to control and monetise their works.
  2. To enable world‑leading AI development in the UK through lawful access to data.
  3. To improve transparency and trust between the creative and AI sectors.

A wide range of options were considered, from licensing‑only approaches to broad text and data mining exceptions. Initially, the Government indicated a preference for an opt‑out model, subject to new transparency obligations.

What the consultation showed

Stakeholder engagement was high. More than 11,500 responses were submitted, with a clear message emerging from creative industries in particular: strong resistance to opt‑out exceptions and broad support for licensing‑based models requiring consent.

The consultation succeeded in surfacing evidence and views. What it did not produce was a consensus on a solution that met all policy objectives simultaneously.

Where things now stand

In its March 2026 Report and Written Ministerial Statement, the Government confirmed it will not proceed with legislative reform at this stage.

The previously favoured opt‑out exception has been set aside. No alternative model has been chosen. Instead, the Government has emphasised the need for further analysis before making changes to copyright law.

That caution is understandable to a certain degree. Copyright reform in the AI context carries real trade‑offs, but the result is that existing law, and its uncertainties, remain firmly in place.

What is still moving forward

The pause on reform does not mean inactivity. The Report highlights continued work on:

  • Transparency around AI training data and web‑scraping practices
  • Labelling of AI‑generated content
  • Technical tools such as metadata and watermarking
  • Exploration of scalable licensing mechanisms

These strands are useful and necessary. However, they sit alongside (rather than resolve) the underlying legal uncertainty around AI training.

Why the current approach remains challenging

The Government’s emphasis on “getting this right” is sensible in principle. But prolonged deferral has consequences. In the absence of clearer rules, risk continues to be allocated unevenly, often to individual creators, SMEs and early‑stage businesses with limited leverage.

Transparency initiatives and market solutions cannot fully substitute for legal clarity.

What clients should take away

  • Rights holders should continue to assume their rights are enforceable and invest in monitoring and licensing strategies.
  • AI developers should plan on the basis that no copyright exception is imminent; data provenance, consent, and transparency remain essential.
  • Boards and investors should continue to treat copyright exposure as both a legal and reputational issue.

The consultation has clarified the difficulty of the problem. It has not yet resolved it.
While further work may be justified, the risk is that delay becomes a policy outcome in itself, leaving uncertainty to harden rather than diminish.

Further information 

If you have any questions regarding this blog, please contact Christopher Perrin in our Corporate, Commercial & Finance team.

about the author

Christopher Perrin is a highly experienced solicitor who leads the Corporate, Commercial and Finance team’s general Commercial & Technology Contracts, Outsourcing & Data legal advisory services.

Contact_us

Let us take it from here

Whatever your legal needs, we’re here to help.

Contact us