Five real, sourced case studies — from a Fortune-verified $700K/year faceless YouTube operation to a $4,843 KDP coloring-book month and an honest $5,000 Etsy loss — each labeled by how well it's actually verified.
This post contains affiliate links. If you buy through them, Aliteq may earn a commission — at no extra cost to you. Prices verified at publish time.
Share
A 22-year-old runs five YouTube channels with no camera, no crew, and AI narration doing the talking — and Fortune reviewed his actual AdSense payout records and found he's clearing something like $700,000 a year. That's not a screenshot in a Discord group or a guru's launch-week claim. A business magazine looked at his real numbers and confirmed them. Stories like that are exactly why "make money with AI" is one of the most searched phrases of 2026 — and also why it's a minefield of fabricated stats and screenshotted lies. So here's the version with actual receipts: five real, sourced examples, labeled honestly by how well they're verified, plus what it would actually take for you to try a smaller version of each one.
I make a living inside this exact economy, so I'm not writing this from the outside. The honest take up front: for every verified $700K/year outlier, there are thousands of people who tried the same playbook and made nothing. Both things are true at once. Let's get into who's actually making what, and how.
1. Adavia Davis: ~$700K/year from faceless AI YouTube — Fortune verified this one
The struggle here is the one every aspiring creator hits: making YouTube videos is slow, and showing your face is a real barrier — for privacy, for confidence, or just because a 22-year-old doesn't have a film crew. Adavia Davis's answer was to stop trying to be on camera at all. He built roughly five faceless channels around AI narration and generated or licensed footage, and instead of treating it as a side project, he ran it like a small production company — testing formats, watching what retained viewers, and scaling the ones that worked. Fortune didn't take his word for the result: reporters reviewed his actual AdSense payout records and channel analytics directly and found he's pulling in roughly $40,000–$60,000 a month at an 85–89% profit margin, because once the pipeline exists there's no crew, no studio, and almost no per-video cost. Annualized, that's in the neighborhood of $700,000. Tier 1 — independently verified by [Fortune](https://fortune.com/2025/12/30/ai-slop-faceless-youtube-accounts-adavia-davis-user-generated-content/). You don't need five channels or his volume to start the same way: pick one niche, script it, and let AI handle the voice and b-roll while you focus on the hook — our faceless YouTube guide walks the exact production line, and Higgsfield is where I generate the b-roll and voiceover for mine.
2. Aitana López: the AI 'model' whose business is worth close to $1M
The struggle here belonged to a small Barcelona design agency, The Clueless, run by Rubén Cruz — projects kept falling through because human models and influencers were unreliable, expensive, or unavailable on short notice. Instead of fighting that, the agency built its own: Aitana López, an entirely AI-generated 'model' with a consistent face across every shoot. She now books brand deals in fashion, fitness, and supplements — reportedly earning on the order of €1,000 per sponsored post, with strong months pushing close to €10,000 — and the wider AI-model business the agency built around her is estimated at $800,000 to $1,000,000. Tier 1 — well-documented, see [Wikipedia's sourced entry](https://en.wikipedia.org/wiki/Aitana_L%C3%B3pez). The mechanism she's built on — one locked, consistent AI face reused across unlimited scenes and outfits instead of a new photoshoot every time — is exactly what our AI influencer guide walks through, using Higgsfield's Soul ID to keep the same face from shoot to shoot.
3. Lil Miquela: the original AI influencer, still earning after a decade
Miquela is the case that proves this isn't a 2026 fad — she's been a fully virtual Instagram persona since 2016, and brands have kept paying for her the entire time: Calvin Klein, Prada, Samsung, and others have run campaigns with a character who doesn't exist. Trade coverage reports she's averaged around $2 million a year since her debut, putting her total brand-deal, music, and merch earnings north of $10 million, built on an audience of roughly 2.6 million Instagram followers. Tier 1 — trade-press reported (use "reportedly"), per [SUCCESS](https://www.success.com/how-ai-influencers-generate-millions-per-year). The takeaway isn't "build the next Miquela" — it's that a consistent AI persona, done well, has now proven it can sustain a real income for nearly a decade, not just a viral month.
The actual setup behind most of these stories: a laptop, a ring light, and months of iteration — not a studio. · Generated with Higgsfield
4. AI coloring books on Amazon KDP: a real, unaudited $4,843/month case study
This is where we drop out of Tier 1 territory, and I want to be upfront about that. The struggle: making a coloring book used to mean paying an illustrator per page or spending a weekend badly tracing clip art, which priced most people out of even trying. One creator publishing on Amazon KDP wrote up exactly what changed for them — using AI image generation to produce clean line-art pages at near-zero cost, then finding an under-served niche instead of a generic "animals coloring book" that nobody searches for. Their own case-study writeup reports $4,843 in a single month from that one niche. Tier 2 — self-reported, unaudited; this is one creator's own claim, not an independently verified figure. Source: Money Tent's case study on Medium. It's also, honestly, the high end — plenty of people publish a coloring book, sell three copies, and quit. If you want to try it, our coloring-books KDP guide has the niche-picking and generation workflow, and it's the cheapest AI generation you can run (line-art models cost fractions of a cent per page).
5. AI print-on-demand on Etsy: one $900 month, and one honest $5,000 loss
Same category, same honesty requirement. The struggle: a decent print-on-demand design used to mean either design skills you didn't have or paying someone who did. One seller with no design background documented generating original art with AI and listing it on Etsy, reporting $912 in their first month through organic search, a few Reddit posts, and a launch promo. Tier 2 — self-reported, unaudited. Source: "Print-on-Demand 2.0" on Medium. But the honest counter-example matters just as much: another creator documented blowing $5,000 on tools, ads, and a course for an AI-powered Etsy shop that never found a niche before the trend they were chasing was already saturated. Tier 2 — self-reported. Source: "I Blew $5,000 on an AI-Powered Etsy Shop" on Medium. Same tools, opposite outcomes — the difference was niche research and restraint on spend, not the AI.
The $900 month and the $5,000 loss both started at a kitchen table with the same tools. · Generated with Higgsfield
Higgsfield · AI creator toolkit
Sponsored
Every AI-native income stream above leans on the same handful of capabilities — a reusable voice, a consistent face, fast image generation, quick video. Higgsfield bundles them in one place instead of six subscriptions. Free tier to test before you commit.
Realistically, yes, in the sense that the tools and the monetization paths above are all real and available to anyone reading this today. Realistically, no, in the sense that you should expect the $912 month, not the $700,000 year, as your honest starting baseline — and even that took someone actually finishing and shipping something. The people in this article who are doing well share one habit: they picked a lane, they iterated in public for months, and they treated the AI part as the cheap, fast part of the job — not the whole job.
HiggsfieldSponsored
Build your own version — start small, verify as you go
Generate the voice, the face, or the visuals for whichever lane above fits you, and treat month one as a test, not a launch. Free tier to try it before you pay for anything.
Is this realistic — how much do most people actually make?
Most people who try any of these lanes make very little, or nothing, especially in month one. The verified $700K/year outlier and the self-reported $900-a-month wins are both real, but they sit at the top of a distribution where the median result is closer to the $5,000-loss story than either of them. Budget for months of unpaid iteration before you judge whether it's working.
Are these numbers actually real, or just marketing?
Each one is labeled by how solid it is. Adavia Davis's figures were independently reviewed by Fortune from his actual payout records — that's as verified as this space gets. Aitana López and Lil Miquela are well-documented in trade press and reference sources but not independently audited line-by-line. The coloring-book and Etsy figures are one creator's own self-reported claim, unaudited — real people, real posts, no outside verification.
Which of these is the cheapest and easiest to start?
AI coloring books on KDP have close to the lowest cost to test — line-art image generation is inexpensive per page, and Amazon handles printing and fulfillment, so you're only risking your time on a first small book. Faceless YouTube has a slower payout timeline (ad-revenue thresholds take months to clear) but compounds better once it's rolling.
Do the AI tools you recommend guarantee results like these?
No, and be suspicious of anyone who implies otherwise. None of the people in this article are Higgsfield customers — we're not claiming that. We recommend it because the tool categories behind every example (AI voice, a consistent face, fast image and video generation) are exactly what these income streams run on, not because using it guarantees any specific income.
These five are real, sourced examples — not the full list of what's possible. I collect the AI money models I've actually tested, with the honest pros and cons, in AI Studio; the e-commerce-specific playbook (product ads, UGC spokespeople, video ads) lives in AI for e-commerce; and the two individual case studies above have their own full walkthroughs if you want to go deeper — faceless YouTube and building a consistent AI influencer.
