Europe’s next move on artificial intelligence and copyright is no longer framed as a single yes-or-no question. A European Commission consultation, updated on October 27 and extended through November 10, places four different policy tools on the table: better information, easier licensing, new remuneration mechanisms and stronger protection against synthetic imitation.
Those options can sound interchangeable when compressed into the phrase “protect creators.” They are not. Each addresses a different failure in the path from an artist’s work to an AI system, and each would distribute responsibility differently.
Information: can creators discover use?
The first problem is visibility. A creator cannot negotiate, object or bring a claim without knowing that a work was used. The EU AI Act already requires providers of general-purpose AI models to maintain copyright policies and publish sufficiently detailed summaries of training content. The consultation asks whether additional measures are needed because the practical information available to rights holders may still be too general to support action.
For visual artists, useful disclosure would need to connect a system’s broad data sources to identifiable works or collections. A list of domains may show where material came from without showing which images were copied, retained or processed.
Licensing: can permission be made workable?
A second path is to improve licensing, potentially through mediation or arbitration. This treats AI training as a use that can be authorized under defined terms rather than as a conflict that must always end in litigation.
The opportunity is a functioning market. The risk is scale. An individual artist cannot negotiate separately with every model provider, and collective systems can obscure how money reaches the people whose work created the value. Any licensing framework will be judged by what it says about scope, attribution, audit rights, downstream models and withdrawal—not simply by the existence of a payment.
Remuneration: who receives the money?
The Commission is also examining mechanisms to ensure appropriate remuneration. Payment is not identical to consent. A statutory or collective payment scheme may compensate uses that creators did not individually approve, while a voluntary license begins with authorization.
That distinction matters for artists whose bodies of work include private, culturally sensitive or politically vulnerable material. A system designed only to distribute revenue may not give a creator meaningful control over whether particular work enters training at all.
Synthetic imitation: what if no file is copied?
The fourth path concerns performers whose voices, appearance or performances are imitated by AI. Copyright can protect a recording, photograph or script, yet a synthetic replica may target identity without reproducing one protected file in an obvious way.
Visual artists face a related question when a system is marketed through recognizable stylistic imitation. Copyright generally protects expression rather than style as an abstract category, so a product can create commercial substitution and public confusion without fitting neatly into a conventional copying claim.
What artists should preserve now
The consultation does not change the law by itself. The Commission has provisionally scheduled a review report and possible legislative proposal for the first quarter of 2027. Until then, creators can strengthen their evidence: preserve dated source files, contracts and publication records; attach accurate authorship and rights metadata; record where high-resolution files are supplied; and keep copies of platform terms that govern uploads.
None of those steps guarantees that an opt-out will be honored or that an unknown use will be discovered. They do establish a clearer record when licensing, attribution or enforcement becomes possible.
The most important outcome may be the refusal to treat one tool as a complete solution. Transparency without a remedy only reveals harm. Licensing without auditability can become a black box. Remuneration without consent can price over refusal. Identity protection without workable enforcement can remain symbolic.
For artists, the policy question is therefore not simply whether Europe will regulate AI training. It is whether creators will be able to know, decide, negotiate, be paid and challenge misuse—and whether those rights can operate together.
ATIM approaches artificial intelligence with curiosity, independence, and balance. We explore its extraordinary potential to empower artists while examining the questions it raises around authorship, copyright, consent, creative identity, transparency, and human expression.
We are neither advocates for AI nor opponents of it. Our focus is how the technology is used—and what that means for artists.
Above all, our commitment is to artists. We look at every development through the lens of what can support, protect, and empower the creative community—identifying opportunities, examining potential risks, and providing artists with the information they need to make informed decisions about their work, their rights, and their future.