Latest / Influencer Marketing with Fexingo: Creators, Sponsorships, and Modern Endorsement Deals / How Influencer Campaigns Are Using Generative AI Avatars
Transcript
- Lucas: So last week, a major luxury fashion house dropped a campaign for their new fragrance — and the face of it wasn't a person. It was a fully ai generated model. No human influencer was paid for that shoot. Luna: I saw that. The backlash was pretty immediate — people called it creepy, dystopian. But also, the engagement numbers were through the roof. So brands are clearly watching. Lucas: Exactly. And that tension is exactly what we're digging into today. This isn't hypothetical anymore. Generative AI avatars are being used in real campaigns, and the economics are shifting fast. Luna: So let's start with the basics. What actually is an AI avatar in this context? Is it like Lil Miquela — the CGI influencer from a few years ago — or something different? Lucas: Good question. Lil Miquela was a pre-rendered CGI character — she was built by a team of artists and animators over months. What's happening now is different: generative AI tools let you create a photorealistic face and body from a text prompt in minutes. You can iterate looks, expressions, outfits — all via software. Lucas: And the cost difference is staggering. A traditional influencer campaign with a mid-tier creator might run you 20 to 50 thousand dollars for a single post. A generative avatar — after the initial setup cost of maybe 5 to 10 grand — costs virtually nothing per use. You can generate hundreds of variations. Luna: Right, so for a brand that wants to test multiple creative directions, that's a no-brainer on paper. But the risk is trust. If a consumer finds out the person in the ad doesn't exist, do they feel manipulated? Lucas: That's the central question. And the answer depends on disclosure. In the luxury fragrance case, the brand did put a small disclaimer in the caption — 'model generated by AI' — but it was easy to miss. Some commenters felt that wasn't enough. Luna: I also think there's a deeper issue: what happens to human models and influencers who rely on those brand deals? If a brand can get the same visual result without paying a person, that's a threat to livelihoods. Lucas: Absolutely. We're seeing it ripple through the industry already. Some influencer agencies are quietly building their own libraries of AI avatars — essentially creating a roster of synthetic talent they can license out. They're not replacing their human creators yet, but they're preparing. Lucas: And then there's the legal side. Who owns the likeness of an AI avatar? If a brand uses a generator that was trained on thousands of real faces, there's a potential right-of-publicity issue. There are already lawsuits percolating around this. Luna: So let's talk about a concrete example. I read about a skincare brand that used an AI avatar for a full Instagram campaign. They didn't disclose it at first — only after a journalist noticed the irises were too symmetrical. Then they added the disclosure. Lucas: That skincare campaign is a great case. They reported a 30 percent higher click-through rate compared to their previous human-influencer campaign. But when the story broke about the lack of disclosure, they saw a spike in negative sentiment. So the performance metrics were good, but the brand trust took a hit. Luna: It seems like the disclosure piece is really the make or break factor. If you're transparent, consumers might accept it as a creative choice. If you hide it, you risk a backlash. Lucas: Exactly. And that's where we're seeing some smart brands experiment. They're creating AI avatars that are explicitly fictional characters — think of them as digital mascots — and they're very upfront about it. The avatar has a name, a backstory, and the brand says 'this is our AI ambassador.' Lucas: That approach seems to avoid the deception pitfall. It becomes a branding asset rather than a replacement for human endorsement. Some automotive brands have done this with virtual spokesmodels for car launches. Luna: So what's the key takeaway for a marketing manager listening right now? Should they be looking into AI avatars for their next campaign? Lucas: I'd say: experiment, but with clear guardrails. First, always disclose. Second, think about whether an AI avatar actually adds to the brand story, or if you're just trying to save money. The cost savings are real, but if the audience feels cheated, you lose more than you gain. Lucas: Third, consider the context. For product demos, explainer videos, or seasonal campaigns where consistency matters, an AI avatar could be perfect. For authenticity-driven categories like wellness or personal finance, human influencers still carry more weight. Luna: Right. So it's not a simple replacement. It's a new tool in the toolkit, but one that comes with its own risks. Lucas: Exactly. And I think we'll see more regulation around this within the next year. The FTC has already signaled interest in synthetic media disclosures. So brands that get ahead of that curve will be better off. Luna: You know, this conversation has actually made me think about my own work. I use generative AI to draft some content ideas, but I never considered using it to create a face for a campaign. It's a whole different ballgame. Lucas: It really is. And if today's episode sparked something useful for you, you know, the way these conversations stay ad-free and independent is through listener support. If you've gotten value from what we do, you can head to buy me a coffee dot com slash fexingo. That's buy me a coffee dot com slash fexingo. Luna: Yeah, it's a simple way to keep this going. We appreciate everyone who's chipped in. Lucas: Alright, back to the avatars. One more angle I think is worth covering: the cultural impact. As AI avatars become more common, how does that affect what we consider 'influence'? Can a fake person really influence real purchasing decisions? Luna: Early data suggests yes. A study from earlier this year showed that consumers aged 18 to 24 were just as likely to trust a recommendation from a clearly labeled AI avatar as from a human influencer — as long as the content was engaging. Lucas: I saw that study too. It's fascinating because it suggests that for younger audiences, the line between real and synthetic is blurrier than we think. The content quality matters more than the humanness of the creator. Luna: Which raises a bigger question: if AI avatars can perform as well as humans, will we eventually see a tier of 'super-influencers' that are entirely synthetic? Brands could theoretically own those IP forever. Lucas: That's the direction some agencies are heading. They're creating avatars with long-term contracts — or rather, perpetual licenses — and the brand never has to worry about a scandal, aging, or renegotiation. It's a controllable asset. Lucas: But that also means those avatars can't evolve organically. They don't have real experiences, real opinions. Over time, that lack of authenticity might become a liability. Audiences are pretty good at sniffing out what's manufactured. Luna: So it's a trade-off: control and efficiency versus authenticity and connection. And right now, we're in the early days where both sides are being tested. Lucas: Exactly. And the smartest move for most brands is to keep a foot in both worlds. Use AI avatars for high-volume, low-stakes content — think banner ads, product shots, social templates. But keep human influencers for the campaigns where trust and storytelling matter most. Luna: That feels like a balanced approach. And it's what I'm hearing from the agencies I talk to — they're not firing their creators, they're supplementing them. Lucas: Right. So to wrap up: generative AI avatars are here, they're being used, and they're effective. But transparency and strategy matter. If brands treat them as a shortcut without thinking about the long-term brand equity, they'll hit the same backlash we saw with that fragrance campaign. Luna: Agreed. It's a tool, not a replacement. Thanks, Lucas. Lucas: Thanks, Luna. See you next time.