A warning from Pitjantjatjara artist and cultural leader Sally Scales has brought a persistent problem into sharper focus: when generative systems imitate Indigenous Australian art, the injury is not limited to copying a visual style. It can strip an image from the family, Country, law and cultural authority that give it meaning.
Reporting published on November 2 describes AI-generated imitations appearing across major online marketplaces, often without attribution and sometimes after material has been taken from art-centre websites. Scales’s argument is direct: a pattern is not a detachable decoration. Its legitimacy may depend on who is permitted to make it, what knowledge it carries and the relationships through which it has been passed down.
Copyright does not contain the whole question
Ordinary copyright law can protect a particular original work, but Indigenous Cultural and Intellectual Property reaches further. It can include communal knowledge, stories, symbols, performance, language and responsibilities that do not fit neatly into a system built around an individual author and a fixed term.
The National Association for the Visual Arts has argued that First Nations artists and communities should decide whether and how their art, cultural material and knowledge are used. That principle is difficult to reconcile with a general opt-out model. A system that collects first and asks communities to discover and object later shifts the burden onto people whose work may already circulate far beyond its original context.
The risks are both cultural and economic. Synthetic imitations can compete with living artists, confuse buyers and circulate false associations between a motif and a community. They may also expose restricted or sacred knowledge. A takedown after an image has been reproduced at scale cannot always repair those harms.
What meaningful consent would require
For artists, consent must be specific enough to answer practical questions. Which works may be used? For what model and purpose? Can outputs imitate an identifiable practice? How long does permission last? Can it be withdrawn? Who receives payment, attribution and reporting? What happens when authority is held collectively rather than by one person?
Transparency is equally important. Artists cannot make informed choices if developers do not disclose training sources, and communities cannot challenge misuse they cannot trace. NAVA has called for consent, attribution and remuneration, stronger transparency around training data and dedicated protection for Indigenous Cultural and Intellectual Property. Creative Australia’s principles similarly place recognition, payment and human creative leadership at the centre of responsible AI policy.
These demands do not amount to rejecting technology. They establish conditions under which new tools might be used without treating culture as an unowned reservoir. An artist may decide that a licensed archive, a community-controlled model or a carefully bounded collaboration serves a creative purpose. Another may decide that particular material must never enter a system. Both decisions require respect.
What artists can do now
Existing protections are incomplete, but documentation still matters. Artists can retain dated high-resolution files, contracts, licensing records and screenshots of suspected misuse. They can state permitted uses in agreements, avoid granting vague rights to future technologies and ask platforms or clients whether submitted material may be used for model training. First Nations artists should seek advice that understands both copyright and Indigenous Cultural and Intellectual Property rather than assuming one substitutes for the other.
For museums, publishers and marketplaces, provenance should apply to digital imagery as seriously as it applies to physical objects. Institutions should know who supplied an image, what rights were granted, whether cultural authority was considered and how synthetic material will be labeled. Removing an artist’s name while preserving a recognizable visual language is not neutral processing; it erases the chain of responsibility that makes attribution meaningful.
The larger lesson is that authenticity is not merely a question of whether pixels were copied. In many Indigenous practices, authority lives in relationships. AI policy that sees only files will miss the very thing it is supposed to protect.
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.