A federal rulemaking debate has moved a central problem of synthetic media from the screen into the courtroom: when a photograph, recording or video is offered as evidence, who must prove that it is real?
At an October 15 meeting in Boston, the U.S. Judicial Conference’s Advisory Committee on Evidence Rules examined a possible new Rule 901(c) for material challenged as a deepfake. The working proposal would use a two-step process. An opponent would first have to present enough information to support a finding that an item may have been fabricated with generative artificial intelligence. If that threshold were met, the party offering the item would then have to show that it is more likely than not authentic.
The proposal is not final. The committee’s own agenda makes clear that it is still asking whether deepfakes are likely to flood courts, whether detection will become harder, whether watermarking systems such as C2PA are sufficient and whether the rule should cover generated text as well as audio and video. Reuters reported that representatives of OpenAI and Meta told the panel that no foolproof detection method exists, while judges cautioned against writing a technology-specific rule before a demonstrated wave of false evidence appears.
Why this matters to artists
Artists, photographers, filmmakers and performers create precisely the kinds of media whose authenticity may later be disputed. A studio portrait can become evidence of authorship. A voice memo can document consent. Behind-the-scenes footage can establish how a work was made. A date-stamped installation photograph can support an insurance, provenance or contract claim.
Generative tools do not merely make false media easier to create. They also create what researchers call the “liar’s dividend”: authentic material can be dismissed as synthetic. A creator may therefore face two opposite risks at once—fabricated content presented as real, and real documentation attacked as fake.
The proposed burden-shifting approach shows why provenance should be treated as part of creative practice rather than as an administrative afterthought. A convincing final image may not contain enough information to establish where it came from. The more valuable a work, commission or identity becomes, the more important the surrounding record is.
A practical authenticity file
For significant works and commissions, creators should preserve original camera files, layered project files, export histories and dated backups. Contracts and releases should identify the people authorized to speak, perform or approve on behalf of a project. When AI is used, a short production note can record the tool, model, date, human decisions and source material involved without disclosing every creative secret.
Publishers and cultural institutions should preserve the path by which media arrived: who supplied it, when it was received, what edits were made and what rights or approvals accompanied it. Repeated screenshots and social-media downloads strip away useful metadata. Whenever possible, verification should begin with the source file and the person responsible for it.
Watermarks and content credentials may help, but the committee’s questions show why no single signal is enough. Absence of a credential does not prove fabrication, and presence of one does not replace editorial judgment. Strong practice combines technical signals with human testimony, chain-of-custody records and consistent publication procedures.
Authenticity is becoming infrastructure
The rulemaking debate is still unsettled, and the small number of reported courtroom incidents argues against panic. Its importance lies elsewhere: institutions that decide what counts as trustworthy are preparing for a world in which seeing and hearing are no longer sufficient tests.
For artists, the lesson is not to turn every studio into a forensic laboratory. It is to recognize that documentation can protect authorship, consent, reputation and value. In the synthetic-media era, an artwork’s story may need to be preserved with almost as much care as the artwork itself.
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.