AI Influencers: The Business Case Is Thinner Than the Hype

AI influencers are computer-generated personas, built with generative imagery and scripted content, that brands deploy on social media in place of, or alongside, human creators. They can post without sleeping, never have a scandal, and will never renegotiate their fee mid-campaign. On paper, that sounds like a procurement officer’s dream. In practice, the business case is considerably more complicated than the technology press would have you believe.

The question worth asking is not whether AI influencers are technically impressive. They are. The question is whether they move product, build brand equity, or do anything a well-briefed human creator could not do better for comparable cost. That is a much harder question to answer honestly.

What Is an AI Influencer, Exactly?

The term gets used loosely, so it is worth being precise. An AI influencer is a synthetic persona, typically rendered using generative AI tools, that maintains a social media presence and is used to promote products or ideas. Some are fully automated. Others are human-operated puppets: a real person controls the narrative and writes the captions, but the face, body, and visual identity are entirely artificial.

Lil Miquela is probably the most cited example. She has millions of followers on Instagram, has appeared in brand campaigns for major fashion labels, and has been doing this since 2016, before generative AI made the production process accessible to anyone with a laptop and a Midjourney subscription. What took a specialist studio years of careful brand-building can now be replicated, at least superficially, in a matter of weeks.

That accessibility is what has changed the conversation. The barrier to creating an AI influencer has collapsed. The barrier to making one commercially useful has not.

Why Brands Are Paying Attention

I have sat across the table from enough CMOs to know what drives interest in new formats. It is rarely pure curiosity. It is usually one of three things: a bad experience with a human influencer, pressure to reduce costs, or a competitor doing something that generated press coverage.

AI influencers speak directly to all three. They do not go off-message. They do not post something catastrophic on a Saturday night that lands in your inbox on Sunday morning. They do not come back six months into a contract asking for a rate increase because their follower count has doubled. For brand safety teams, that is genuinely appealing.

The cost argument is also real, up to a point. Top-tier human influencers command fees that would have seemed absurd a decade ago, and mid-tier creators have followed the pricing upward. If you can generate comparable reach with a synthetic persona at lower cost, the spreadsheet looks attractive. The problem is that the cost comparison is rarely done honestly. Production, persona management, content scripting, legal review for disclosure compliance, and the creative labour of keeping a fictional character coherent across months of posting all add up. The all-in cost of a well-run AI influencer programme is higher than most brands expect going in.

If you want a broader view of where AI is reshaping marketing operations, the AI Marketing hub at The Marketing Juice covers the strategic and practical dimensions across channels and use cases.

The Engagement Numbers Are Not What They Seem

Early in my career, I learned a lesson about vanity metrics that has stayed with me. I was running a paid search campaign at lastminute.com for a music festival. We generated six figures of revenue in roughly a day from a relatively straightforward campaign. The numbers were clear: clicks led to bookings, bookings led to revenue. The feedback loop was tight and honest.

Influencer marketing, even human influencer marketing, has never had that clarity. AI influencer marketing has it even less. When a synthetic persona generates high engagement, you have to ask what is driving it. Curiosity is a powerful motivator. People follow AI influencers partly because they find the concept fascinating, partly because the content is algorithmically optimised, and partly because the visual quality is often genuinely striking. None of those motivations map cleanly to purchase intent.

Likes and follows from audiences who are engaging with the novelty of the format are not the same as the trust-based relationship that makes human influencer marketing work at its best. A creator who has built an audience over years, who reviews products honestly and occasionally says something is not worth buying, has credibility that a synthetic persona cannot replicate. That credibility is what converts.

Ahrefs has covered the intersection of AI tools and content performance in useful depth, including practical frameworks for evaluating what AI-generated content actually delivers versus what it promises on paper.

The Transparency Problem Is Not Going Away

There is a regulatory dimension to this that brands are underestimating. Disclosure requirements for influencer marketing have been tightening for years. The FTC in the United States, the ASA in the UK, and equivalent bodies across Europe and Asia-Pacific have all moved toward stricter requirements for sponsored content. AI-generated personas add a layer of complexity that regulators are actively working through.

The core question is whether audiences have the right to know they are interacting with a synthetic persona, not just a sponsored one. In several markets, the answer is moving toward yes. Brands that have built campaigns around AI influencers without clear disclosure are sitting on reputational risk that will not stay dormant indefinitely.

There is also a consumer trust dimension that sits outside the regulatory framework. Audiences who discover they have been engaging with a fictional persona without knowing it tend to feel deceived, regardless of whether any rule was technically broken. That reaction is not irrational. It is a reasonable response to finding out that the parasocial relationship they thought they were building was constructed on a false premise.

I have seen this dynamic play out in other contexts. When I was running agency teams, we had clients who wanted to use fake reviews, manufactured testimonials, or stock photography presented as real customers. The short-term appeal was always the same: it looks better, it costs less, it avoids the messiness of real people. The long-term risk was always the same too. Authenticity failures compound. They do not stay contained.

Where the Technology Actually Has Merit

I want to be precise here, because dismissing AI influencer technology entirely would be as lazy as uncritically embracing it. There are genuine use cases where synthetic personas make commercial sense.

The strongest case is in categories where the product is the hero, not the person promoting it. Gaming, fashion, beauty, and certain technology categories have audiences that are comfortable with constructed aesthetics and are less dependent on the authentic-human-recommendation dynamic. If your target audience already engages with fictional characters, virtual avatars, and constructed online identities as a normal part of their media diet, a well-executed AI influencer is less of a conceptual leap.

There is also a legitimate use case in markets where finding credible human creators is genuinely difficult. Niche B2B categories, highly regulated industries, and markets where the talent pool is thin can benefit from synthetic personas that maintain consistent messaging without the unpredictability of human partners.

Content volume is another honest advantage. A brand that needs to maintain presence across multiple platforms, in multiple languages, at a posting frequency that no single human creator could sustain, has a real production problem that AI tools can help solve. Moz has explored how generative AI imagery is changing content production workflows, and the practical implications for visual content at scale are worth understanding for any team thinking about this space.

The mistake is treating these use cases as proof that AI influencers are categorically superior to human creators. They are not. They are different tools with different trade-offs, and the right choice depends entirely on what you are trying to achieve.

The Brand Safety Argument Has a Flip Side

The brand safety case for AI influencers is real but incomplete. Yes, a synthetic persona cannot be photographed doing something embarrassing. It cannot give an interview that contradicts your brand values, or publicly support a political position that alienates half your customer base. The downside risk from individual talent behaviour is eliminated.

But brand safety risk does not disappear. It relocates.

The risk of using AI influencers is that the brand itself becomes the story, and not in a good way. If a journalist or a sharp-eyed consumer identifies that your “real person” is synthetic, the coverage tends to focus on the deception rather than the product. That is a brand safety problem of a different kind, and it is one that the traditional influencer risk frameworks do not account for.

There is also platform risk. Social media platforms are still developing their policies on AI-generated content and synthetic personas. What is permitted today may be restricted or require explicit labelling tomorrow. Brands that have built significant campaign infrastructure around AI influencers are exposed to policy changes in a way that human creator relationships are not.

HubSpot has done useful work on the cybersecurity and trust dimensions of generative AI, and the risks that brands often overlook when deploying AI-generated content are directly relevant to anyone considering AI influencer programmes at scale.

What the Production Reality Looks Like

Early in my career, I was told no when I asked for budget to build a website. Rather than accept that as the final answer, I taught myself to code and built it. The point of that story is not that resourcefulness always wins, although it often does. The point is that understanding how something is actually built changes your assessment of what it costs and what it is worth.

Most brand teams commissioning AI influencer programmes have not spent time understanding the production reality. They see the polished output and assume the process is straightforward. It is not.

Maintaining a coherent synthetic persona at posting frequency requires significant creative oversight. The visual consistency alone, making sure the persona looks like the same person across hundreds of images generated over months, requires careful prompt engineering, style guides, and quality control that most generative AI tools do not handle automatically. The narrative consistency, keeping the persona’s backstory, opinions, and relationships coherent over time, requires the kind of character development work that is closer to fiction writing than to marketing operations.

Then there is the audience management layer. Comments, DMs, and community interactions need responses that are consistent with the persona’s voice. Brands that automate this entirely tend to produce interactions that feel mechanical and undermine the whole premise. Brands that use human operators to manage the interactions are, at that point, running an elaborate puppet show that requires more human labour than they anticipated.

Semrush has covered AI content strategy in practical terms, and their analysis of how AI tools fit into content workflows gives a grounded view of where automation adds genuine value versus where human oversight remains essential.

How to Evaluate Whether This Makes Sense for Your Brand

Having managed marketing budgets across more than 30 industries, I have seen enough shiny objects come and go to know that the right framework for evaluating any new channel or format is the same: start with the business problem, not the technology.

If you are considering AI influencers, the questions worth answering honestly before committing budget are these.

What specific business outcome are you trying to drive, and why is an AI influencer the right mechanism for it? If the answer involves phrases like “test and learn” or “stay ahead of the curve,” that is not a business case. That is curiosity dressed up as strategy.

How does your target audience relate to authenticity in this category? If your customers make purchase decisions based on trusted recommendations from real people, a synthetic persona is working against the mechanism that makes influencer marketing effective. If your customers engage primarily with content aesthetics and entertainment value, the calculus is different.

What is the all-in cost, including production, legal review, platform management, and the creative overhead of persona maintenance? Compare that honestly to what a well-briefed human creator programme would cost for equivalent reach and frequency.

What is your disclosure strategy, and does it meet current regulatory requirements in every market you are operating in? This is not optional. It is the foundation on which everything else sits.

Ahrefs has explored how AI is changing SEO and content strategy in ways that are relevant to anyone building a content programme around synthetic personas, including how search visibility is affected by AI-generated content at scale.

The Honest Assessment

AI influencers are a genuinely interesting development in the influencer marketing space. They are not a revolution. They are a new tool with a specific set of advantages and a specific set of risks, and the brands that will use them well are the ones that are clear-eyed about both.

The hype cycle around AI influencers follows the same pattern I have watched play out across every major technology shift in marketing over the past two decades. The technology arrives, the early adopters generate press coverage, the coverage attracts followers who do not fully understand the underlying mechanics, and then the market sorts itself out into the use cases that actually work and the ones that were always more interesting as a concept than as a commercial proposition.

We are somewhere in the middle of that cycle right now. The brands that will come out ahead are not the ones that move fastest. They are the ones that ask the right questions before they move at all.

HubSpot’s overview of AI marketing automation is worth reading alongside any evaluation of AI influencer programmes, because the underlying question, where does automation add value and where does it subtract it, is the same in both cases.

Moz has also done useful thinking on how AI tools fit into broader content and SEO strategy, which is relevant context for any brand thinking about AI-generated content at scale, including the influencer channel.

For more on how AI is reshaping marketing strategy across channels, the AI Marketing section of The Marketing Juice covers the practical and strategic dimensions without the hype.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions