AI influencers: What consumer sentiment data tells brands

As you scroll through your TikTok or Instagram feed, you’ll no doubt come across at least a handful of influencers trying to sell you something by wearing the latest viral lip gloss, hopping into luxury cars, and taste-testing popular treats. But what if I told you that while these content creators may look like real people, they’re actually AI influencers?

If you’re a marketer or business owner who is already fully using AI in your digital marketing strategy, then you’re likely already familiar with the concept. You may even be curious about leveraging AI influencers in your influencer marketing strategy.

But how will your target consumers feel about the use of AI influencers? Will they be receptive, or can AI influencers erode the trust you’ve built with your audience? Can AI influencers really save you money compared to working with human influencers? And are all virtual influencers AI? Keep reading for answers to these questions and more.

Table of Contents

• What are AI influencers, and why are brands considering them?
• What do consumers think about AI influencers and AI models?
• Do AI influencers build or erode brand trust?
• Notable Virtual Influencers + Brand Examples to Learn From
• Do AI influencers save brands money?
• How to Use AI-Assisted Content Workflows Without Losing Authenticity
• Common Pitfalls With AI Influencers and How to Avoid Backlash
• What’s Next for AI Influencers
• Frequently Asked Questions About AI Influencers

What are AI influencers, and why are brands considering them?

AI influencers are virtual personas whose appearance, voice, behaviors, or content are created or assisted by generative AI. An example of an AI influencer is Aitana Lopez, who has over 400,000 followers on Instagram. She’s marketed as a virtual fitness and lifestyle influencer and has promoted brands such as Amazon, Victoria’s Secret, Olaplex, and Razer.

Her work as an influencer has reportedly earned her and The Clueless, the Barcelona-based tech company that created her, over $11,000 per month.

AI influencers fall under the umbrella of “virtual influencers,” but make no mistake — while every AI influencer is a virtual influencer, not all virtual influencers are AI-generated. For example, the virtual influencer, singer, and model Miquela Sousa (formerly known as Lil Miquela) was created in 2016 using artist-driven CGI, and her music is created by a collective of singers, songwriters, and instrumentalists—not generative AI.

Over the last decade, Miquela Sousa has worked with brands like Gucci, Prada, and Calvin Klein. She’s also the first virtual avatar to sign with a model agency, having signed to CAA in 2020 and previously being represented by WME.

Why are brands interested in AI Influencers?

Brands are dabbling with using AI influencers and virtual models like Aitana Lopez and Miquela Sousa for three reasons — creative control, reduced costs, and scalable content.

Creative Control

Ruben Cruz, founder of The Clueless, says Aitana Lopez was created so the company could make a “better living and not be dependent on other people who have egos” or “who just want to make a lot of money by posing.”

In other words, Cruz wanted to avoid working with influencers whose personalities, creative processes, and work styles clash with his company’s creative vision—a sentiment brands wanting to leverage AI influencers can relate to.

AI and virtual creators will recite a script, deliver a video, or conform to a brand’s aesthetic exactly as agreed upon. Companies never have to worry about unwanted input or script changes that could slow production.

Reduced Costs

The initial creation of an AI influencer or virtual internet personality can cost anywhere from a few hundred dollars to a few hundred thousand. However, once out of the initial setup phase, AI influencers can save companies money on lighting, production, and paid partnerships.

Scalable Content

Human influencers need to sleep, eat, take breaks, pick up their kids from school, or walk the dog — they can’t make content 24 hours a day, 7 days a week. But AI influencers can. Though AI influencers like Aitana Lopez may look super busy doing their makeup, traveling, and hopping into luxury cars, it’s all fake— so they always have time to make content.

What do consumers think about AI influencers and AI models?

Many consumers feel that AI influencers corrode trust, with one Reddit user saying, “If I see something that is AI but tries to appear real, it’s an instant turn-off and trust killer. It’s enough for me to see it once for a brand, and the trust is gone forever.”

The skepticism toward AI influencers shows up in surveys, too. According to a study published via ResearchGate, 45.3% of consumers would view an ad with an AI influencer without taking action, 28.3% would ignore the ad completely, and only 26.4% would actively engage with the ad, such as by clicking or searching.

Do AI influencers build or erode brand trust?

AI influencers pose a high risk of eroding trust between brands and consumers because using a fake person to promote a brand’s goods or services doesn’t read as authentic. The appeal of human influencers and content creators is that they’re real people who cultivated a consistent following by being their genuine selves.

Influencers and content creators bond with their followers over shared interests, hobbies, and beliefs, and they often get their audience interested in a product or service by using it themselves.

In fact, a survey by Clutch shows that if a human influencer is found to use AI-generated content, 71% of consumers say they would question the recommendation, and 53% are less likely to buy from a brand known for using AI in social content.

An AI or virtual influencer can’t use a product or service, can’t have a genuine interest, nor a human experience to connect with human followers.

That said, the trust risk is not equal in every situation. It concentrates where audiences are skeptical, where safeguards are missing, where disclosure is weak, and where nobody is measuring the effect. Score each area below before you decide to leverage AI influencers.

Audience fit

Your audience already engages with virtual or synthetic creators, and the category doesn’t rely on personal experience (e.g., entertainment, fashion visuals, gaming). You have research or past results showing acceptance.

Audience attitudes are unknown, or the fit is plausible but untested. Run a small survey or pilot first.

Recommendations depend on lived experience (skincare results, health, finances, parenting), or research shows that your audience is skeptical of AI personas.

Brand safety

The persona has written guardrails (topics, tone, claims), tiered approvals, a named person who can pull content immediately, and documented rights to the persona and any likeness used.

Guardrails exist, but approvals, ownership, or escalation paths are incomplete.

Posting or replies are fully automated, ownership is unclear, or the persona comments on social issues without sign-off.

Disclosure readiness

The AI label appears in the bio and on every relevant post, paid content is marked as an ad, and legal has approved wording for each market.

Templates exist but haven’t been reviewed for every market and platform.

The plan is to present the persona as a real person, to disclose that the persona hasn’t been reviewed, or to run ads in New York or the EU without checking current requirements.

Measurement

You have a human-creator benchmark, defined KPIs (conversion, sentiment, brand trust), a window long enough to see past the novelty effect, and a pre-set stop threshold.

KPIs are defined but there’s no benchmark, or the window is too short to read results.

You’d rely on follower counts or likes alone, with no sentiment tracking and no stop criteria.

How to score it:

• Any No-go: Don’t launch. Fix the issue or choose a different approach.
• All Go: Proceed to a limited pilot.
• Mix of Go and Conditional: Launch only if each Conditional item has an owner and a fix date before the first post.

Why these four areas matter: Consumers aren’t rejecting AI across the board. That same Clutch survey also found that 42% view AI positively when it’s used for educational content, but 64% would trust a brand or creator less if AI were used to misrepresent products. Nearly 90% say it’s important to disclose AI-generated content.

Disclosure is also becoming a legal matter. For example, as of June 9, 2026, New York requires certain ads featuring AI-generated human performers to include a disclosure, and the EU’s Article 50 transparency obligations have applied since August 2, 2026. Always double-check to confirm what applies to your markets.

Notable Virtual Influencers + Brand Examples to Learn From

Virtual influencers have been working with brands for years, so their campaigns show what works and what doesn’t. Each example below is broken down by what it teaches about creativity, community, transparency, representation, and governance.

1. Miquela x PAC SUN Back-To-School Campaign

PacSun teamed up with Miquela for its back-to-school and holiday 2022 campaigns, and said the deal marked the start of a broader push into virtual influencers. The two collaborated on social content to amplify those seasonal campaigns after both had taken part in the ComplexLand digital event.

Creative

The partnership was built around a persona who already lived online, with roughly 3 million Instagram followers at the time. Choosing a virtual star with an existing audience meant the brand didn’t have to build one from scratch.

Community

The fit rested on shared audience territory. PacSun framed the move as a way to strengthen its connection with a core demographic exploring the digital space. Takeaway: pick a persona whose community already overlaps with yours.

Transparency

Miquela is open about being a virtual influencer, describing herself as a “Robot living in LA,” so there was no deception to manage. The audience was in on it from the start.

Representation

PacSun’s president pointed to Miquela’s involvement in the Black Lives Matter movement and the Rally the Vote campaign as aligned with the brand’s values. That raises a fair question for any brand: a persona’s advocacy is authored by its creators, so scrutinize whose voice is really speaking.

Governance

A virtual star never ages, clothes always fit perfectly, and she does what you ask, but that control is also a responsibility. Set clear approval processes for any social or political messaging attached to a persona, since you are effectively the speaker.

2. Samsung and Sam

Sam is Samsung’s own virtual influencer, created in collaboration with Cheil and the Brazilian design firm Lightfarm. She is used mainly in Latin America, and in July 2023, she connected to the regional launch events from Seoul, where she covered the global Galaxy Unpacked event.

In January 2024, she was named Samsung’s AI ambassador ahead of Galaxy Unpacked 2024, where she was set to present products and interview five live influencers from Latin America.

Creative

Sam works because she’s integrated into real brand moments rather than confined to social posts. She has blogged, made tutorials, released a music video promoting the Galaxy Z Flip 5, and helped sell devices at Unpacked events.

Giving a persona several jobs, from host to educator to performer, builds her into the brand’s story rather than treating her as a one-off stunt.

Community

Samsung paired Sam with real people. At Unpacked 2024, she interviewed Latin American creators live, bridging the virtual and human sides of the community. Her role was also regionally focused, which suggests a persona can work better as a local community figure than as a global one.

Transparency

Sam is openly presented as a virtual influencer, and Samsung’s own newsroom calls her one. Her role is also clear (host, guide, ambassador), so audiences aren’t asked to believe she personally uses Samsung products.

Representation

As a brand-owned character, Sam’s look, voice, and personality are entirely Samsung’s choices. Treat that as a casting decision: involve regional teams and creators in shaping a persona who will speak to their audience, rather than designing her centrally and exporting her.

Governance

Samsung reported that Sam helped boost early sales of its new devices in Latin America, including a 30% increase in sales on its Mexico live-shopping portal.

Set measurable goals and a content plan before launch, attribute any performance claims to their sources, and decide in advance who will maintain the persona, since a neglected virtual influencer can quickly feel stale.

3. Amazon Prime, boAt & Kyra

Kyra was launched by FUTR Studios in January 2022 as a virtual influencer that is both CGI and AI-powered. Her first brand partner was boAt for the #FutureOfAudio campaign, which launched its Rockerz 330 and 333 ANC neckbands.

Amazon Prime Video followed, integrating Kyra into the promotion for its sci-fi series “The Peripheral,” with her creator noting that the team recreated the show’s VR headset and transported her to its setting.

Creative

Both campaigns leaned into science fiction, which is where a virtual persona excels. boAt said the collaboration enabled a sci-fi-like audio-visual experience that wouldn’t be possible with any other influencer. Choose briefs a human can’t easily deliver.

Community

The boAt launch reached beyond social posts. Kyra appeared at an in-person boAt party through a 3D projection installation, alongside human creators. Bridging virtual and live moments helps a persona feel part of a scene rather than an ad.

Transparency

The pairing with Prime Video worked partly because it was an obvious fiction-meets-fiction match. When the story is openly synthetic, disclosure becomes part of the creative rather than a caveat.

Representation

Kyra was positioned as India’s first virtual influencer, and her creators said they planned a new model with a South Indian focus. That’s a useful prompt for brands: local relevance matters, and a single persona rarely represents an entire market.

Governance

Early results were strong, but they come from launch-novelty periods. Judge a persona on sustained performance, not a debut.

4. Dior & Noonoouri

In 2018, Dior’s beauty team had Noonoouri recreate its Rouge Dior Ultra Rouge launch film. The original starred Natalie Portman; the tribute was nearly a shot-for-shot reenactment, down to the Dior ring. Noonoouri also took over the Maison’s Instagram during the Dior Cruise show.

Creative

Rebuilding a famous campaign with a stylized animated character created instant buzz. The contrast between the original and the tribute did most of the work.

Community

Dior used the persona to reach younger Instagram audiences interested in high-end beauty. The takeaway is to match a persona to an audience you’re trying to grow, not just one you already have.

Transparency

Noonoouri’s doll-like proportions make her unmistakably an animation, so nobody is misled about what they’re seeing. A clearly stylized persona carries less deception risk than a photorealistic one.

Representation

Her exaggerated features reflect a stylized ideal, not a real body. Brands should consider what beauty standard a digital character promotes, especially in cosmetics.

Governance

Stepping into a role originally played by a real actress raises questions about likeness, permission, and optics. Clear agreements around any reference to a real person’s work are essential.

Do AI influencers save brands money?

AI can save brands money, but it depends on what you count. AI influencers can cut the most visible line item, production, while adding costs elsewhere: oversight, disclosure, rework, reputation, and measurement. A fair comparison looks at the total cost of the program, rather than the cost of a single post.

You can look up costs online but be aware that cost claims are mostly published by vendors and agencies that sell AI content, and they vary widely. Some guides puts AI influencer creation at , with monthly costs of $500 to $15,000+. But other sources claim 80-95% savings on large volumes of video. Treat all of these numbers as starting points for your own testing, not benchmarks.

Taking a Total-Cost View

1. Production. This is where savings are most credible. Once a persona exists, the marginal cost of producing one more piece of content approaches zero, and there are no talent fees, travel, or scheduling.

The catch is the upfront build and the ongoing work of generating, editing, and approving content. A persona that looks effortless usually still has a team behind it.

2. Risk management. Brands often cite reduced risk of scandal as a draw. In a World Federation of Advertisers finding reported by Favikon, 58% of brands interested in AI influencers named reduced risk of influencer scandals as a top draw. But the risk doesn’t disappear completely; it only changes shape.

You swap the risk of a creator’s behavior for the risk of your own content errors, off-brand outputs, or a backlash against the concept itself. Budget for review and an escalation plan.

3. Disclosure compliance. This cost is growing. Platforms like Instagram and TikTok have their own AI labeling rules and protocols. And other countries and regions have legal requirements regarding AI disclosure. In the EU, for example, transparency obligations for AI-generated content began to apply in August 2026.

The direct cost is modest, covering labels, templates, and legal review. The hidden cost is audience trust: consumer distrust is a recognized downside when disclosure is unclear.

4. Rework. Generated content needs corrections: anatomical glitches, inconsistent facial expressions, misreading cultural cues, tone issues. Rework hours rarely appear in a vendor’s cost comparison, so track them in your own use case.

5. PR and reputation. As I said earlier, public perception of AI influencers or synthetic personas ranges from curiosity to hostility, and it varies by category and audience. Set aside time and budget for a communications plan, including how you’ll respond if people accuse you of replacing real creators.

6. Community impact. Human creators bring a ready audience and a relationship with it. A new AI persona has to build both. The savings can also be offset on the other side: some AI adopters found that when creators use AI to cut their own costs, those savings haven’t led to lower fees, because brands pay for influence and credibility.

And partnerships with AI accounts aren’t a sure thing; Data shows that brand deals with AI social accounts fell by about 30% in the first eight months of 2025 compared with the same period in 2024.

7. Measurement. Fairly comparing AI and human influencers takes planning. You need matched campaigns, consistent metrics, and enough time to see past the novelty effect. Some of this has a real cost in analyst hours and tooling, and without it, you’re making decisions based on anecdotes.

As analysis puts it, the honest comparison isn’t a cost debate but a question of what a campaign asks the influencer to do.

Where AI Influencers Tend to Save Money (and Where They Don’t)

They tend to help when you need high volumes of content, many variations to test, a tightly controlled brand look, or multilingual output. They tend to be a poor fit when the sale depends on lived experience or personal credibility, such as skincare results, health claims, parenting, or financial advice.

There, the cheaper post can cost you in conversion and trust.

A Qualitative Worksheet to Complete Before You Model Any Finances

Before you build a spreadsheet, answer these in plain language. If your team can’t answer them, then you’ll already know you need to reevaluate your strategy.

Production

What would we make in a typical month, and who would create, edit, and approve it? Who owns the persona long term?

Risk management

What could go wrong with our own content (errors, off-brand posts, offensive output)? Who has authority to pull something fast?

Disclosure compliance

Which markets and platforms are we in? What labels are required? Who signs off on wording?

Rework

How often do we expect to fix or redo outputs? What would “good enough” look like?

What’s the likeliest criticism? Who speaks for us, and what would we say?

Community impact

Which human creators or communities does this affect? Are we replacing them or adding to them?

Measurement

What would success look like in 90 days? What human-led benchmark are we comparing against?

Then add two summary questions: What is the strongest reason this would save us money, and what is the strongest reason it wouldn’t? Count the “High” ratings. A cluster of them in risk, disclosure, or community impact is a signal to slow down, no matter how good the production savings look.

Then, build the model.

Once you’ve done the worksheet, a simple model has four parts: one-time setup costs, recurring monthly costs, the hidden costs from your worksheet (rework hours, legal review, monitoring, crisis reserve), and the performance side (conversions, engagement, or sales per dollar spent). Run it against a human-creator pilot of similar scope, and revisit it after a quarter, not a week.

The Bottom Line

AI influencers can lower the cost of making content. Whether they lower the cost of the program depends on your category, your audience, your compliance exposure, and how honestly you count what’s outside the production line. The brands most likely to come out ahead are the ones that measure the whole picture and start small.

How to Use AI-Assisted Content Workflows Without Losing Authenticity

Audiences forgive AI-assisted content that’s honest and consistent. They don’t forgive content that feels off-brand, evasive, or unreviewed. Build the workflow around human judgment, and aim for consistency over perfection.

Codify your brand voice.

Write the voice down in a form both people and AI tools can use.

• Voice guide: Tone, vocabulary, phrases to use and avoid, and a few on-brand and off-brand examples.
• Persona sheet (if you use a virtual influencer): Backstory, values, topics they will and won’t discuss, and visual style.
• Standing prompts: Turn the guide into reusable instructions so every draft starts from the same baseline.

Set clear approvals.

Not every asset needs the same scrutiny. Tier them by risk:

• Low risk (routine posts, repurposed content): Social lead approves.
• High risk (health, finance, political or social topics, anything involving a real person’s likeness): Brand, legal, and social all approve.

Name one person who can pull a post immediately.

Standardize disclosure.

Create templates in advance so disclosure is never improvised. Cover three cases: AI-generated persona, AI-assisted content from a human creator, and paid partnerships. Each should state who made it and whether it’s an ad, in plain language. Have legal review the templates once, then reuse them.

Check requirements for each platform and market, since they differ and are changing.

Govern your assets.

Keep one source of truth for approved content.

• Store approved images, prompts, character references, and disclosure templates in a single library with version history.
• Log which tool and prompt produced each asset.
• Record usage rights, especially for any real person’s likeness or voice.
• Retire outdated assets so they can’t be reused by accident.

A content management system, such as HubSpot Content Hub, can serve as a shared home for assets and approval steps, but the process matters more than the tool.

Run review loops.

• Before publishing: Check voice, accuracy, cultural fit, and disclosure. Reject anything that looks polished but feels hollow.
• After publishing: Review comments and sentiment weekly in the early weeks.
• Monthly: Compare a sample of recent posts against the voice guide to catch drift, and update the guide with what you learn.

Establish who owns what.

Brand

Voice guide, persona sheet, final say on what’s on-brand

Legal

Social

Favor consistency over realism.