AI Has Turned Standard Creator Contracts Into “A Legal Minefield”

For years, likeness clauses in creator agreements were largely boilerplate. A brand would secure rights to use campaign assets across social channels, paid media or websites for a defined period of time. But generative AI has dramatically expanded what those assets can become.

Now, creators, agencies and brands are confronting a far more complicated question: who owns an AI version of a person?

The urgency is being accelerated by widespread AI adoption inside marketing departments. According to the IAB’s Creator Economy Ad Spend & Strategy Report 2025, three in four brands are already using, or planning to use, AI for creator marketing-related tasks, signaling how quickly AI has moved from experimentation into operational infrastructure.

Thomas Markland, founder of creator agency HYDP, says the industry entered this territory much faster than most executives anticipated.

“I think AI cloning capabilities have moved faster than most of the industry anticipated,” Markland said in an interview. “Last year, likeness and usage clauses in creator contracts were mainly boilerplate and just a formality.”

That is no longer true. “Now, these clauses are increasingly scrutinized as new AI tools for creators rapidly develop,” he said. “Typically, we’d obtain full rights to the content produced; however, this level of ownership opens up potential options for brands and agencies to utilize and manipulate assets using artificial intelligence as they see fit.”

In practice, that means a creator who filmed a single campaign months ago could theoretically have their likeness repurposed indefinitely across AI-generated campaigns they never physically participated in.

According to Markland, contracts have suddenly become significantly more complex.“Contracts have quickly become a legal minefield, the complexity of which is exacerbated by the rapid pace of AI development,” he said.

The Rise Of “Kill Switches” And AI Usage Clauses

The stakes are rising because brands increasingly want perpetual rights and maximum flexibility over creator assets, while creators are beginning to realize how much future value may be embedded in their face, voice and behavior.

“Brands want unlimited, perpetual use,” Markland said. “Some creators are pushing back hard, and more will follow, demanding time-limited licences, approval rights over every AI-generated output and revenue participation clauses tied to usage volume.”

One of the most notable developments, he says, is the emergence of so-called “kill switches.”

“We're also seeing kill switches become more common–contractual provisions that allow a creator to revoke their likeness licence if the brand uses the clone in a context the creator hasn't approved.”

That kind of language would have seemed excessive in influencer contracts only a few years ago. Today, it reflects growing concern over what happens when AI-generated content scales beyond direct human oversight.

The uncertainty extends beyond future campaigns. Markland says one of the biggest unresolved questions involves historical creator content that was produced before AI cloning became commercially viable.

“While most creators and agents are slowly becoming familiar with these new clauses, from a legal standpoint, it’s interesting to ponder what brands could do with historically produced assets that lack these stipulations,” he said.

What Exactly Is A Creator Licensing?

The issue becomes even more complex when defining what exactly is being licensed in an AI deal. Is it a face? A voice? A personality? A behavioral pattern?

“The question of where the line is regarding biometric data or likeness is one the industry hasn’t been able to answer yet, and frankly, neither have the courts,” Markland said.

He believes many companies entering the AI creator space are still underestimating both the legal ambiguity and the reputational risk involved.

“Most brands are relatively new to the world of AI content and are deeply concerned about consumers' perceptions of AI use in marketing communications,” he said. “That said, as AI advances, in practice, brands will be attempting to purchase everything in a single transaction, face, voice, mannerisms, even what I'd describe as a behavioral signature, the way a creator delivers a line, their cadence, their catchphrases.”

The commercial appeal is obvious. A creator could theoretically license their likeness once, allowing brands to generate campaigns at scale without repeated shoots, travel or production costs.

“In this new age, a brand could feasibly partner with a creator without the creator ever having to film a video or attend an event,” Markland said. “For those with established followings, they simply license their characteristics and let the brand produce and distribute communications on their behalf.”

AI Cloning Could Create Winners And Casualties

For top-tier creators, that could become a highly lucrative business model. Markland argues that AI cloning may actually strengthen negotiating leverage for the biggest names in the creator economy.

“For the mega influencer tier, the Khaby Lames, the MrBeasts, the creators with tens of millions of deeply loyal subscribers, AI cloning is genuinely strengthening their position,” he said. “Their likeness has become a licensable IP asset in the same way a Hollywood actor's does.”

But he believes the effect could be far more destabilizing for mid-tier creators. “For mid-tier creators, the picture is more complicated and frankly more worrying,” he said.

“A brand that previously needed ten mid-tier creators to cover a campaign can now use one, license their likeness, clone it into ten variants and eliminate nine ongoing relationships.”

That dynamic could reshape the economics of influencer marketing altogether. “The creators who should be most alert to this are those whose value to brands has historically been reach and content volume rather than a deeply irreplaceable personal brand,” Markland said.

“If your value proposition is that you make good content efficiently, an AI clone makes that argument for you and then makes you redundant.”

The Trust Problem Hasn’t Gone Away

Despite the excitement surrounding AI-generated content, consumer skepticism remains significant.

A 2025 Vogue Business AI Consumer Perception Survey found that more than seven in ten respondents said they would never trust an AI influencer, while only a small minority trusted AI-generated recommendations over human creators.

That tension is becoming one of the defining contradictions of the AI creator economy. Brands are rapidly adopting AI tools while consumers still overwhelmingly value human authenticity.

Markland believes the distinction lies in the difference between fully synthetic influencers and AI-enhanced human creators. “The key distinction is that the audience already has a trust relationship with the human creator; the AI is extending that relationship, not manufacturing it from scratch,” he said.

He points to localization as one of the clearest commercial use cases.

“Localization is a clear example of where the hybrid content model is apt–a creator with an English-speaking audience can now have their likeness deliver the same campaign in Portuguese, German, and Japanese, with cultural nuance baked in, without setting foot in a studio.”

The Definition Of A Creator Is Changing

The implications extend beyond marketing efficiency into questions of IP ownership, consent and labor economics.

“Traditionally, creators have been in control of their image,” Markland explained. “If they see their image or video in a place they did not realize was being used, contracts can be revised, or they could pursue legal action.”

“However, adding these new dynamics with AI cloning in the mix, creators may not even know the image or video being produced; they never stood for a photo shoot or recorded the video, it just pops up as part of a wider deal.”

For now, regulation has not kept pace. “Until we have legislation that treats biometric data the way GDPR treats personal data in the EU, with explicit purpose limitation and consent requirements, creators are signing away far more than they realize,” Markland said.

Meanwhile, brands themselves are still trying to navigate the operational and reputational implications of AI-generated creator content. CreatorIQ’s State of Creator Marketing Report 2025-2026 found that AI governance, authenticity and brand safety are rapidly emerging as central concerns for marketers integrating AI into creator campaigns.

The creator economy has historically rewarded reach and consistency. AI may increasingly reward ownership, trust and contractual control. “If a creator's likeness can generate content autonomously, the definition of a creator shifts from someone who makes things to someone who owns a content identity,” Markland said.

He believes the creators best positioned for the AI era will be those with the strongest audience relationships. “The most valuable creators of the next decade won't necessarily be the most talented or the most prolific,” he said. “They'll be the ones who built the deepest audience trust before AI made replication trivial.”

As brands race to scale AI-generated creator content, the next major battle in influencer marketing may not be over distribution or monetization, but over who ultimately owns a creator’s digital identity.