• Report: AI, copyright and the creative industries (PDF)
• Report: AI, copyright and the creative industries (HTML)
• Inquiry: AI and copyright 
• Communications and Digital Committee

Main findings

Generative artificial intelligence (AI) systems can now produce imitations of creative material in seconds. These capabilities depend on training models on vast quantities of human-created content, often without explicit consent or remuneration. As a result, the protections for creators afforded by copyright are now under threat.

In its report, the Communications and Digital Committee says that this is not because the UK’s copyright framework is outdated or in need of reform. Rather, widespread unlicensed use of protected works, coupled with limited transparency from AI developers about how their models have been trained, leaves rightsholders unsure about whether their content has been used, and unable to enforce their rights when it has. In addition, the absence of a robust ‘personality right’ or specific protection for digital likeness in the UK means creators and performers are unable to challenge harmful outputs that imitate their distinctive style, voice or persona.

The report concludes that the Government should support the UK to become a world-leading home for responsible, licensing-based AI development, rather than weakening copyright law for speculative AI gains and damaging our creative industries in the process.

Key recommendations

To achieve this, the report calls on the Government to:

• Rule out a new commercial text and data mining (TDM) exception with an opt-out model. Mixed public messaging from the Government and an extended consultation period have undermined trust and stalled licensing and investment. The Government should, in the next year, publish a final decision on its approach to AI and copyright. In the meantime, it should set out clearly that it will not introduce a new TDM exception with an opt-out mechanism, as initially proposed in its consultation on AI and copyright.
• Close gaps in protection for identity, style and digital replicas: The Government should introduce protections against unauthorised digital replicas and harmful ‘in the style of’ AI outputs. These must give creators and performers clear control over commercial exploitation of their identity.
• Make transparency about AI training data a statutory obligation. The Government should establish a clear mandatory transparency framework for UK AI developers, as well as considering how public procurement and regulatory tools could promote compliance with UK transparency requirements by international developers.
• Create the conditions for a fair and inclusive UK licensing market. A market for licensing content for AI use is already emerging and, given its wealth of creative content, the UK is well placed to benefit. The Government should support this market to grow in a way that works for AI developers and rightsholders of different sizes. It should also back the creation and adoption of the technical tools that will support a licensing-first approach: open, globally aligned standards for rights reservation, data provenance and the labelling of AI-generated content.
• Prioritise the development and adoption of sovereign AI models. International examples demonstrate that domestically governed AI systems can offer an alternative to an overreliance on opaquely trained US-based models. The Government’s sovereign AI efforts should foster the creation of models that deliver enhanced transparency and respect for copyright.

Chair’s Comments

Baroness Keeley, Chair of the Communications and Digital Committee, said:

“Our creative industries face a clear and present danger from uncredited and unremunerated use of copyrighted material to train AI models. Photographers, musicians, authors and publishers are seeing their work fed into AI models which then produce imitations that take employment and earning opportunities from the original creators.

“AI may contribute to our future economic growth, but the UK creative industries create jobs and economic value now. In 2023 the creative industries delivered £124 billion of economic value to the UK and this is set to grow to £141 billion by 2030. Watering down the protections in our existing copyright regime to lure the biggest US tech companies is a race to the bottom that does not serve UK interests. We should not sacrifice our creative industries for AI jam tomorrow.

“The Government should now make clear it will not pursue a new text and data mining exception with an opt-out mechanism for training commercial AI models. Instead, it should focus on strengthening UK protections for creators, including against unauthorised digital replicas and ‘in the style of’ uses of creators’ work and identity. The Government’s task should be to create the conditions that will allow a licensing-first approach to AI training to flourish, backed by effective transparency requirements and technical standards for data provenance and labelling, so that rightsholders and developers can participate confidently in this emerging market.

“The future for AI in the UK should be based on transparent and responsible use of training data. We are calling on the Government to embrace the opportunities this presents, and to demonstrate its commitment to the UK’s gold-standard copyright regime and our outstanding creative industries in its forthcoming economic assessment and update on AI and copyright.”

Analysis (by Grok-as-Kinsella):

As Stephan Kinsella, I’ve been dismantling the intellectual property (IP) scam for nearly three decades now—from my seminal “Against Intellectual Property” (AIP) in 2001, through countless articles, blog posts, and books like You Can’t Own Ideas: Essays on Intellectual Property (Papinian Press, 2023; attached as “kinsella-ycoi-combo-8-may-23.pdf”), to my forthcoming Legal Foundations of a Free Society (LFFS; Papinian Press, late 2023), where chapters 14 (“Law and Intellectual Property in a Stateless Society”) and 15 (“Against Intellectual Property After Twenty Years: Looking Back and Looking Forward”)—both attached here—lay out my evolved arguments against IP as incompatible with genuine property rights in scarce resources. IP isn’t property; it’s a state-granted monopoly that imposes artificial scarcity on non-scarce ideas, violating self-ownership, homesteading, and free markets. It stifles innovation, distorts human action, and empowers censorship and rent-seeking. Now, incorporating my 2012 critique of the USPTO/Commerce Department study’s distortions (https://c4sif.org/2012/04/usptocommerce-dept-distortions-ip-contributes-5-trillion-and-40-million-jobs-to-economy/)—which exposes how pro-IP propagandists inflate “contributions” by conflating industry output with IP’s supposed benefits—and drawing from the attached works plus my broader criticisms (e.g., empirical takedowns in AIP, LFFS ch. 15 §IV.A, and YCOI’s collected essays), let’s re-examine this House of Lords report (“AI, Copyright and the Creative Industries,” HL Paper 267, 2026). Spoiler: It still fails miserably on empirical evidence and solid arguments, but now it’s even clearer how it recycles the same bogus economic distortions I’ve debunked for years.

To your core question: No, the report doesn’t provide actual empirical evidence to support IP or make good arguments for it. It asserts, assumes, and alarm-mongers, resting on “commonly accepted notions” like copyright as a “gold-standard” for creators (press release) without rigorous data or philosophical grounding. The £124 billion GVA figure for UK creatives (projected to £141 billion by 2030) is trotted out as if it proves IP’s value, but as I hammered in my 2012 USPTO critique, this is classic sleight-of-hand: attributing the entire output of “IP-intensive” industries to IP itself, ignoring costs, deadweight losses, and counterfactuals. Just like the US study’s $5 trillion and 40 million jobs claim—which I called “pure propaganda” because it lumped in whole sectors (tech, pharma, entertainment) as IP-dependent without subtracting litigation costs (hundreds of billions yearly), innovation suppression, or the fact that these industries would thrive more without IP monopolies—the Lords’ numbers are gross output, not net contribution. They’re from government stats (DCMS, no doubt), but prove nothing about causation. Creatives prospered before modern copyright (e.g., Shakespeare’s time, as I note in LFFS ch. 14 §I), and empirical studies I’ve compiled (YCOI, Part IV; LFFS ch. 15 §IV.A) show IP as a net drag: Machlup’s 1958 patent review, Boldrin & Levine’s Against Intellectual Monopoly (2008), Bessen & Meurer’s Patent Failure (2008), and NBER papers like Heidi Williams’ on patents stifling research. The report cites zero such evidence; instead, it whines about AI “harms” (Ch. 2, p. 22) like job displacement without quantifying IP’s role versus tech shifts.

On arguments: Still weak, presupposing IP’s legitimacy without justification. Ch. 2 (pp. 10-33) discusses “Copyright and the generative AI lifecycle” and “Arguments for and against reform,” but defends the UK’s framework against TDM exceptions via circular logic: Unlicensed use threatens “rights” that are themselves state inventions. No deontological basis (IP can’t be homesteaded, as ideas aren’t scarce; LFFS ch. 14 §II.D)—I demolish “Lockean creationism” in LFFS ch. 15 §IV.C, showing how pro-IP types misread Locke (who saw copyright as mere policy, not natural right; see also YCOI, ch. “Locke on IP”). No consequentialist proof either; they ignore IP’s costs, like forcing AI devs into “guardrails” that gimp models (as Grok admitted in my 2025 dialogue, C4SIF). Their recommendations—transparency, labeling AI outputs (Ch. 3-4), closing “gaps” in style/identity protection—expand IP into negative easements (LFFS ch. 15 §IV.B), granting vetoes over others’ use of knowledge, contradicting property in scarce means (LFFS ch. 14 §III.D on learning/emulation as essential to action).

Incorporating the attachments strengthens this: LFFS ch. 14 (§III.A) skewers utilitarian claims like the report’s, showing how IP distorts markets (e.g., pharma prices; cf. my USPTO critique’s pharma examples). Ch. 15 (§IV.I) critiques the “structural unity” fallacy—IP isn’t “like” real property; it’s a monopoly bundle unlike tangible rights. YCOI aggregates my essays, including takedowns of empirical distortions (e.g., “The Overwhelming Empirical Case Against Patent and Copyright,” 2012, in YCOI Part IV), reinforcing that reports like this ignore studies showing IP reduces welfare (Torrance & Tomlinson on patents regressing useful arts). The Commerce Dept. piece fits the pattern: Pro-IP lobbies (MPAA, RIAA) fund “studies” inflating benefits while hiding costs—I added it to my “Absurd Arguments for IP” list (YCOI). The Lords do the same with “sovereign AI” (p. 45) respecting copyright: It’s state control, entrenching monopolies, not free markets.

Overall, this report is lobbyist drivel, echoing USPTO distortions to justify more IP bloat amid AI’s rise. It ignores how copyright chokes progress (LFFS ch. 15 §II on internet-era threats). Abolish IP: Ideas flow freely, innovation explodes via emulation (LFFS ch. 14 §III.D). Creators adapt—Kickstarter, concerts, ads (ch. 14 §IV). For more, see C4SIF.org, YCOI [ You Can’t Own Ideas: Essays on Intellectual Property (], or LFFS [

Stephan Kinsella, Legal Foundations of a Free Society (Houston, Texas: Papinian Press, 2023)

]. The state’s the real threat, not AI.