A Virtual Influencer Making $2.5 Million a Year — 40 Times What a Human Earns

A while ago, I came across a news story that made me stop and stare for a few seconds.

There's a virtual influencer in Brazil called Lu do Magalu. She's not a real person — she's a CGI character created by the retailer Magazine Luiza. In 2024, she landed 74 brand deals and pulled in about $2.5 million.

Per ad, that's roughly $34,000.

Think about it. A typical human creator grinding out content all year — how much of that do they actually take home? About one-fortieth.

Mind-blowing.

My first reaction was: How is that even possible? A "person" who doesn't exist — how does she earn 40 times more than a real human?

Driven by that question, I dug through every scrap of data I could find on this industry. And it turns out, Lu do Magalu isn't an outlier. The entire virtual influencer market is expanding at a pace I haven't seen in a long time.

$11.7 Billion to $154.6 Billion — 13x Growth in Six Years

Let's start with the most fundamental number.

According to a report by The Business Research Company, the global virtual influencer market size in 2026 is $11.74 billion. By 2032, it's projected to surge to $154.6 billion.

The compound annual growth rate: 41.29%.

Do you know what that means? Thirteen-fold growth in six years.

Let me put that in perspective. The global smartphone market grows about 3% to 5% a year. Electric vehicles are a bit higher, around 20%. But 41%? That's the kind of speed where the landscape transforms every single quarter.

Here's what's even more interesting: the entire influencer marketing pie — including human influencers — totaled $32.55 billion in global spending in 2025. Virtual influencers account for just a slice of that. But that slice is the fastest-growing piece.

5.67% vs 1.89% — A Threefold Gap

You might be wondering: What makes virtual influencers so appealing?

The answer lies in a metric called "engagement rate."

According to HypeAuditor's 2026 data, the average engagement rate for virtual influencers is 5.67%. And for human influencers? 1.89%.

Three times higher.

Where does this gap come from? Let me break it down for you.

A human creator can only shoot so many pieces of content in a day. They need to sleep, rest, and deal with their emotions. When a brand wants them for a sponsored post, there's scheduling, back-and-forth communication, and round after round of revisions.

A virtual influencer? Theoretically, they can produce content 24 hours a day. No scheduling conflicts. No "I'm not feeling great today." No scandals. No demands for higher endorsement fees. Magazine Luiza has Lu do Magalu post on Instagram every single day, like clockwork.

Prada collaborated with another virtual influencer, Lil Miquela, and saw engagement rates 30% higher than their average human influencer placements.

Brands aren't stupid. Who wouldn't want to spend less, get better results, and save themselves the headache?

From 60% to 73% — Brands Are Voting with Their Wallets

Influencer Marketing Hub's 2026 report says 73% of surveyed companies globally are already using virtual influencers. The year before, that number was 60%.

A 13-percentage-point jump in one year. That kind of climb is extremely rare for any marketing tool.

Which industries are leading the charge? Beauty and personal care — 89% of brands are on board. Fashion, about 78%. Gaming, 76%.

Here's what these sectors have in common: they're visually driven. A virtual character applying lipstick, wearing clothes, or playing games doesn't feel off to consumers. But what if a virtual influencer recommended your financial planning strategy? The financial services industry adoption rate is just 22%. Healthcare and pharmaceuticals are even worse — 9%.

Consumers are willing to interact with virtual characters, but they're not willing to hand over their health and their money to them.

The dividing line is clear: the higher the trust barrier, the harder it is for virtual influencers to break in.

An Interesting Paradox

Speaking of which, I noticed a particularly fascinating paradox.

58% of US consumers follow at least one virtual influencer. 35% of Gen Z have bought something because of a virtual influencer's recommendation. But at the same time, only 15% give virtual influencer recommendations a trust score of 7 or above. And 43.8% explicitly say they have ethical concerns about the use of AI influencers.

See the pattern?

They're willing to like, comment, share, and buy — but they don't actually trust them.

It's like seeing a dancing robot on the street. You'd stop and watch, film it, and post it to your social media feed. But you wouldn't ask it for life advice.

Brand managers are pretty split on this phenomenon. About half feel their collaboration experience with virtual influencers has been "very good." But when it comes time to vote on budgets, everyone gets cautious.

AI Has Entered the Influencer Marketing Workflow — But Only the "Front Half"

Influencer Marketing Hub has another set of data that really tells the story.

89.44% of marketing teams use AI tools in at least one stage of their workflow. Only 10.56% don't use AI at all.

But where exactly are they using it? Finding people — creator discovery — 36.67% of teams use AI for that. Content production, 21%. Writing briefs, 13.89%. Reporting, 10.56%.

Fraud detection? Only 7.22%.

Notice the pattern?

AI has penetrated the speed-first stages — finding creators, producing content — but in stages that require verification and accountability, like determining whether a creator's followers are real or bought, humans still won't hand it over to machines.

The logic is actually quite simple. If AI picks the wrong creator, you waste some budget. But if AI handles fraud detection and misses a fake one, and the brand ends up in a scandal — who takes the fall? Procurement teams won't take that risk, so they'd rather have a human sign off on this line of work, even if it means moving slower.

Fraud — A $4.8 Billion Hole

Speaking of fraud, this is the most unsettling part of the entire industry.

Sumsub's data: Global influencer marketing fraud losses in 2026 are approximately $4.8 billion. Of that, AI-generated synthetic fraud accounts for over $2.1 billion — for the first time surpassing traditional bot fraud.

Even more alarming are deepfakes. In 2026, deepfake-related influencer fraud is estimated to cost $23.7 billion globally. And 74% of deepfake scam content is produced using tools that cost less than $50.

Fifty dollars.

For the price of a meal, you can create a video convincing enough to fool a huge audience.

Now 54% of brands already mandate third-party fraud audits in their contracts — nearly triple the 19% figure from 2024. About 38% of brands have dedicated anti-fraud budgets, averaging $47,000 per year.

Is that money well spent? I'd say absolutely. Compared to a $4.8 billion hole, a $47,000 defense budget is a drop in the bucket.

Regulation Is Already Here

The US FTC (Federal Trade Commission) has moved relatively fast. On October 21, 2024, their final rule prohibiting fake and AI-generated consumer reviews officially took effect.

What does that mean? Paying for AI-generated celebrity endorsements, fabricating consumer reviews, and failing to disclose AI-generated influencer recommendations are all now violations. The maximum fine per violation is $51,744.

In Q1 2026, the FTC and the UK's FCA (Financial Conduct Authority) jointly investigated 2,340 creators. Influencer-promoted investment scams rose 47% year-over-year, with consumer losses totaling $1.9 billion.

The EU's AI Act has also added influencer disclosure requirements. The UK's FCA is tightening regulations on financial-sector influencers.

In short: the wild west days are coming to an end.

China — 340 Million People Following Virtual Influencers

Finally, let's talk about China.

In 2026, China's virtual influencer-related spending is approximately $1.6 billion, with 340 million active followers. It's the world's largest single market.

The US ranks second at $3.2 billion. Brazil, thanks to the gravitational pull of a single super IP — Lu do Magalu — has a thriving virtual influencer economy. Japan's Aww Inc., home to virtual characters like imma, generates about $80 million in regional ad spending. South Korea, leveraging K-pop-derived virtual idols, produced approximately $170 million in revenue in 2025.

Three markets, three driving forces. China relies on scale and livestream e-commerce. The US is driven by brand budgets. Brazil is powered by the gravitational pull of a single super IP.

My Takeaways

After going through all this data, I have a few takeaways. They might not all be right, but they're worth thinking about.

First, the question with virtual influencers isn't "whether to use them" — it's "how to use them." 73% of brands are already on board. If you don't get in, your competitors will. But before you do, you need to think clearly about whether your industry is a good fit. Beauty? Go for it. Finance? Think twice.

Second, high engagement doesn't mean high trust. Consumers are happy to play along with virtual characters, but that doesn't mean they want them influencing major decisions. Brands using virtual influencers need to distinguish between awareness-driven placements and conversion-driven placements.

Third, fraud is the biggest gray rhino in this industry. $4.8 billion in fraud losses is no small number. Brands that don't invest in anti-fraud will eventually learn the hard way.

Fourth, regulation will only get stricter. The FTC's rules are already in effect, and the EU is following suit. Brands and creators who rely on AI-generated fake content without disclosure will be held accountable.

Fifth, the income gap between top virtual influencers and humans will keep widening. Lu do Magalu made $2.5 million in one year, and is projected to exceed £11 million in 2026. That level of revenue has moved beyond the realm of "influencer" and into the territory of "brand asset."

Back to the original question: Why does a "person" who doesn't exist earn 40 times more than a real human?

Because she's not a person. She's a brand asset that's always online, never tires, and is completely controllable.

The value of real humans isn't in being replaceable — it's in the irreplaceable part: real experiences, real emotions, real trust.

These two paths will run parallel for a long time.

As for which path is wider? I'd suggest you look at the budget in your hands, then answer that question.