A practical field guide for building, launching, growing, and monetizing characters that live between media, software, and community.
Campaigns, performances & the people behind them
New research / September 2026
20 historical cases and 25 current research guides. Follow the source, read the analysis, and take a practical question into your next brief.
Browse research & archive ↗
Research guideProduction · Reviewed · 19 September 2026
The right character format follows the recurring performance, revision and delivery burden—not the most impressive demo in the pitch deck.
Read the guide
Research guideBusiness · Reviewed · 19 September 2026
A virtual character makes production look flexible. The useful brief still fixes the claims, files, approvals, reuse permissions and exit conditions before anyone renders the first frame.
Read the guide
Research guideGrowth · Reviewed · 19 September 2026
A narrow pilot tests whether a specific audience returns for a repeatable editorial job before the operator commits to a bigger lore, production or commerce system.
Read the guide
From the archivePlatforms · Event · 4 August 2025
The August 2025 launch positioned AI characters not only as chat partners, but as units of short-form, participatory entertainment.
Read the case
From the archiveProduction · Event · 2 July 2025
A July 2025 image drop made replicas of real models part of fashion marketing, while labor advocates pressed questions about control and work.
Read the case
From the archiveProduction · Event · 24 April 2025
The 2025 Style Coach activation moved a virtual model from broadcast imagery to real-time, natural-language retail interaction.
Read the casePut the research to work
Choose the decision in front of you: a production method, an audience promise, or a commercial brief.
Start with the audience and the role the character plays. Then choose the visual system, performance rules and pilot that can test the idea.
4 pieces · Follow the path ↗02 / Build the systemA polished image is a deliverable, not a production system. These guides connect source assets, human performance, voice, animation and review.
14 pieces · Follow the path ↗03 / Publish with a purposeSeparate attention from repeat interest. Design a content promise, measure comparable cohorts and give people a safe reason to come back.
8 pieces · Follow the path ↗04 / Model the exchangeA sponsor buys a defined use, not a follower count. Work through deliverables, rights, contribution margins and repeatable revenue before scaling output.
9 pieces · Follow the path ↗The most durable projects do not sell AI images. They build a recognizable point of view, a repeatable production system, and a place for fans to matter.
Hatsune Miku, Ami Yamato and pre-social avatar culture prove that a character can carry music, performance, and fandom.
Miquela, Kizuna AI, Shudu and Imma turn the feed into a stage for identity, fashion, and serialized narrative.
VTubers, live chat, motion capture, and platform-native communities make the operator invisible but the relationship immediate.
Character.AI launches a social feed; Aww pairs imma with an LLM; new studios sell consistent identities rather than one-off images.
The edge moves to disclosure, machine-readable provenance, owned audiences, licensing, and characters that can act in real time.
Earlier supplied briefing · Reviewed 15 September 2026. The new research library contains individually sourced updates reviewed 16 September 2026; it does not revalidate every estimate in this earlier summary.
New field notes from the supplied September 2026 brief. The useful question is not whether a character is “AI”; it is which production, audience, rights, and revenue system it belongs to.
AI-native characters can move from concept to first post in days with a small software stack. But consistency, story, disclosure, moderation, sales, and rights still determine whether the account survives beyond the novelty cycle.
Source reviewed 15 Sep 2026 Read the supplied research brief ↗
Reference use only: these are published visual frames for design and editorial context, not cleared commercial artwork. Open the source and rights ledger ↗
Organize by the work the character must do. Each archetype implies a different pipeline, moat, and monetization mix.
Shudu, imma, Rozy. Highest craft and luxury credibility; strongest fit for fashion, beauty, travel, and licensing.
Any Malu, Noonoouri, Anna Cattish. Lower asset cost, easier video continuity, and better merch portability.
Lu, Barbie, GEICO Gecko, Qai Qai. Highest continuity and commercial control; glamour is not the point.
Human performers behind 2D or 3D avatars. Schedule, chat, tips, subscriptions, and community culture drive retention.
Hatsune Miku, K/DA, APOKI, Plave. Music, streaming, concerts, merch, and licensing replace the sponsored-post loop.
Guggimon, Janky, Nobody Sausage, John Pork. Lower uncanny-valley risk and strong merchandise attachment.
Aitana, Yang Mun, and thousands of smaller accounts. Fastest iteration, highest consistency and disclosure pressure.
A user avatar, deepfake, hologram, or customer-service digital human may be adjacent technology without being an influencer business.
Indicative planning ranges from the supplied brief. These are not quotes or market averages.
| Track | Build / monthly range | Time to first post | Core stack | Best fit |
|---|---|---|---|---|
| AI-native | $50–$500/mo | Days | Reference conditioning, LoRA, ComfyUI, image/video models, voice, lip-sync | Solo creators, fast tests, niche media |
| VTuber | $0–$7,000 commissioned | 2 hours to 8 weeks | VRoid or Live2D, VTube Studio / VSeeFace, webcam, OBS | Live community, tips, memberships |
| 2D animation | $10k–$75k build | 2–4 months | Illustration, After Effects, Toon Boom, Moho, or Rive | Video-first, family-safe IP, merch |
| Hyperreal 3D | $100k–$500k+ legacy build | 4–12 months | MetaHuman / Blender / Maya, mocap, facial capture, Unreal or offline render | Luxury, events, licensing, enterprise |
Reference conditioning is fast. LoRA training raises control. A 3D base plus AI finishing gives the most durable identity at the highest effort.
Track Nano Banana, Imagen, FLUX, Seedream, GPT Image, Qwen, Midjourney, Ideogram, and Recraft by model version, not by generic “AI generator” rankings.
Veo, Kling, Seedance, Wan, Runway, Luma, Hailuo, and open lip-sync pipelines compete on continuity, audio, control, and cost per keeper.
Use the range, not the headline. The brief finds a roughly 20× spread in 2026 market estimates because firms define the market differently.
| Signal | Reported figure | Label | Strategic read |
|---|---|---|---|
| 2026 market estimates | $0.77B–$15.9B | Estimate spread | Publish the source and definition; do not collapse incompatible categories into one TAM. |
| Lu do Magalu | ~$2.54M / 74 sponsored posts | Third-party estimate | Brand-owned distribution and media spend can create economics that independent characters cannot copy. |
| Lil Miquela | ~$73.9k sponsored-post estimate; ~$30M funding | Estimate / reported funding | Cultural fame, venture funding, and sponsored-post yield can move in different directions. |
| Aitana López | ~€3k/mo typical; up to €10k strong month | Press-reported estimate | A narrow niche plus a repeatable production system can outperform a broad “lifestyle” pitch. |
| Yang Mun | $213k gross over three months | Creator-reported claim | Interesting proof of the AI-native upside; verify the offer, traffic, costs, and health-content risk. |
| Fanvue | 20% platform cut | Platform economics | Direct-to-fan revenue is attractive but concentrated platform exposure is a single-point failure. |
The supplied brief flags Sora’s 2026 shutdown, OnlyFans’ synthetic-persona ban, and YouTube’s inauthentic-content policy as different versions of the same risk: your character should not depend on one model, one subscription platform, or one distribution channel.
Practical rule: keep source assets, scripts, trained adapters, rights logs, customer data, and at least three revenue lines under your control.
The new brief sharpens the existing guardrails. This summary is editorial research, not legal advice.
The brief identifies Article 50 transparency obligations, machine-readable marking, and disclosure duties for deepfakes and AI-backed direct interaction. Check the European Commission’s current guidance and local implementation.
FTC endorsement rules still require clear, conspicuous disclosure. Digital-replica, publicity, voice, and performer-consent rules vary by state and contract. Do not use a real person’s likeness or voice without written permission.
TikTok, YouTube, and Meta increasingly expose AI labels and provenance signals. Keep a human approval queue for claims, health, finance, minors, politics, and anything that could be mistaken for a real customer.
The brief’s website strategy points to gaps around the established directory model.
Organize by archetype with a current tool stack, cost range, workflow, and case study instead of another alphabetical gallery.
Separate verified earnings, third-party estimates, creator claims, platform fees, and modeled scenarios. The range is the product.
Make disclosure, provenance, consent, and brand safety a maintained product layer for creators, agencies, and buyers.
Editorial note: this update synthesizes the supplied virtual-influencers-research.md file, reviewed 15 Sep 2026, alongside the existing VirtualHumans.org research record. Recheck changing claims at the point of use.
Use the category to choose your production stack and your business model. The more live the character, the more the moat shifts from visuals to operations.
Digital models such as Shudu, Imma and Rozy. Best for fashion, beauty, travel and product worlds.
Real-time bodies and facial rigs turn a performer’s choices into a responsive avatar.
Bold, stylized characters travel well across stickers, animation, games, toys and merch.
Any Malu and brand mascots show the path from social video to television, music and transmedia.
Human performers operate 2D or 3D avatars. Personality, schedule and chat culture drive the loop.
Prompted or fine-tuned characters made for short-form content. Consistency and disclosure are the hard parts.
Characters exist as agents, feeds, scenes, or apps. Their content is something audiences can continue or remix.
Audio, talking-head video, podcasts and live translation make the character useful beyond the feed.
A creator or brand remains visible while AI expands the world, output, localization or customer experience.
This snapshot audits the current VirtualHumans.org index and adds newer AI-native projects. Search by name, handle, or type. Audience numbers are dynamic and should be re-verified before a deal.
Audit note: VirtualHumans.org says 1,000+ fans are required for inclusion and that VTubers are largely not included. The directory below preserves the current index as a research snapshot, not a live ranking.
Image note: roster thumbnails are local copies of public profile assets from VirtualHumans.org and the listed Instagram profiles for identification. Original profile links and image rights remain with their respective owners; reviewed 14 Sep 2026.
Each profile separates public facts from operating analysis. Read these as patterns to borrow, not recipes to copy.
Lu / retail
Miquela / narrative
Shudu / luxury
imma / interface
Aitana / lean stackAny Malu / animationCodeMiko / live
Kizuna / music
APOKI / voice
Nobody / meme
Granny / AI video
Tilly / AI-nativeReference use only: these local profile assets identify public characters and are not cleared commercial artwork. Replace with licensed press images, commissioned art, or rights-cleared recreations before launch. Source context: VirtualHumans.org and linked public profiles.
Lu of Magalu — the owned brand voiceLu is Magazine Luiza’s branded digital character: a helpful, stylish retail host who can appear in content, campaigns, live TV, music videos, and partner work. Her power is being an owned and repeatable brand asset.
Short social videos, product education, lifestyle posts, campaign appearances and cross-platform brand storytelling. Ogilvy describes her as a face for the retailer and its wider presence and value.
Reported partnerships include Adidas/Farm and Magalu’s own commerce ecosystem. Axios estimated roughly $2.5M per year in sponsored posts in June 2026; treat this as an external estimate, not audited financials.
Own the character when the business is really selling continuity: customer trust, product discovery, localization, and an always-on brand voice.
Miquela — narrative before automationCreated by Brud in 2016, Miquela arrived as a fictional Brazilian-American singer and model with a first-person feed. Early growth was choreographed through mystery, editorial coverage, fashion partnerships, music, and a storyline that eventually made the construction visible.
Instagram-first fashion imagery, music releases, short video, culture commentary, philanthropy and collaborations with human talent. Newhouse reports 0 to 1.4M Instagram followers in one year and a Prada partnership.
Publicly documented brand work includes Prada, Calvin Klein with Bella Hadid, Samsung and Pacsun. Newer coverage reports the account remains in the millions; do not use old follower counts as current facts.
A compelling character has a reason to post. Backstory, stakes and continuity create the why now that makes an asset worth following.
Shudu — the asset can outlast the postShudu is The Diigitals’ original digital supermodel, created in 2017. The studio positions her as a Vogue cover star and a continuing collaborator with luxury fashion houses.
Editorial imagery, beauty collaborations, campaign stills, fashion films and virtual runway concepts. Her work is image-led and intentionally premium.
Publicly documented work includes Christian Louboutin, Balmain, Vogue Australia, Lexus and Samsung. The official studio page does not publish audited audience or revenue figures.
If the character is a high-end creative asset, distribution is only one value path. A tight licensing and production workflow can matter more than daily posting.
imma — from model to AI interfaceAww’s pink-haired imma combines a photoreal 3D identity with a cultural and design point of view. In 2024, Aww announced imma AI, an LLM-powered form developed with NVIDIA and designed for more natural conversation and lip-sync.
Fashion and lifestyle imagery, brand campaigns, public appearances, event hosting and interactive virtual human experiences. Aww listed Instagram, TikTok, Weibo, Zepeto, Facebook, Douyin and RED in 2025.
Aww reported 1.5M worldwide followers and work with Amazon, Coach, Porsche and IKEA in June 2024. Aww also reported Instagram above 390k and TikTok above 480k around Expo 2025.
The next layer is not more images. It is a character that can explain, host, localize and respond without losing identity.
Aitana López — the lean AI studio modelThe Clueless created Aitana as a pink-haired, fitness and lifestyle-oriented virtual soul, after identifying demand for a controllable model that could appear across campaigns without travel, scheduling or traditional production overhead.
Instagram-first lifestyle scenes, fitness and fashion content, brand placements, event and media appearances, and paid creator-platform distribution.
The Clueless lists Audi, Amazon, L’Oréal, Coca-Cola, Vogue and other clients. TIME reported in 2024 that the team estimated typical monthly earnings near €3k, with an upside of €10k; Fast Company reported a combined following near 400k and an 11-person team in May 2026.
A consistent identity is a production system. The margin comes from reusable references, prompt templates, approvals, and a clear commercial niche.
Any Malu — a character that escaped the feedAny Malu began as an animated YouTube personality and became the star of a Cartoon Network show. The format is a useful reminder that social characters can graduate into owned media rather than staying trapped in platform metrics.
Animated comedy, backstage content, YouTube episodes, television, collaborations and platform-native sketches.
VirtualHumans’ profile reports 3M+ YouTube subscribers, 280M+ video views and 1M+ collective fans across other networks in its review snapshot. These numbers are historical and dynamic, so verify current channel analytics.
Build formats that can be licensed. A repeatable show bible creates more optionality than an endless stream of disconnected posts.
CodeMiko — participation is the productCodeMiko is a real-time 3D VTuber operated by The Technician. The character’s edge is a live environment where chat can influence the show, interviews, games and scene state.
Just Chatting streams, interviews, interactive games, clips, community drops and multi-platform distribution through Twitch, YouTube, TikTok, Instagram and Discord.
The official Linktree lists Twitch, Kick, YouTube, TikTok, Snapchat, Discord and the Mikoverse. A third-party tracker showed roughly 892k Twitch followers in September 2026; use the platform counter for current campaign diligence.
The expensive part of live virtual talent is not the avatar. It is the showrunner, moderation, interaction design, uptime, and community ritual.
Kizuna AI — category creation through consistencyKizuna AI’s official biography says she began in December 2016 with the theme I want to connect with you, and helped establish the modern VTuber category through an anime-style 3D avatar, a performer, YouTube, gaming and music.
Video, gaming, music, live events, brand and tourism collaborations, fan translation and community participation.
The official site lists 3M+ YouTube subscribers and 10M+ social followers, plus a 2025 return after an activity pause. Her business model spans music, merchandise, events, brand work and character IP.
A category-defining character can turn a technical format into a cultural label. Consistency compounds into an ecosystem.
APOKI — voice-first IP with a visual worldAPOKI began on YouTube and TikTok in 2019, debuted as a virtual K-pop artist in 2021, and built around songs, performance, an identifiable rabbit-like design, and a production team at AFUN Interactive.
Dance and song covers, original singles, music videos, live streams, interviews and fan shout-outs. The music creates a repeatable reason to return.
The official site reports 5M+ social followers and highlights work with Sony Honda Mobility’s AFEELA prototype, E-40 and global producers. Those are company-reported metrics and partnership credits.
Voice and music can make a virtual character feel authored. The visual layer becomes a distribution system for songs, shows, events and licensing.
Nobody Sausage — the short-form loopNobody Sausage is an elastic 3D animated character built for short-form humor, movement, remix culture and visual repetition. The simple silhouette makes the character recognizable at feed speed.
Short-form animation, memes, music, collaborations and platform-native loops. Its global distribution is stronger than its narrative complexity.
A third-party creator database snapshot in September 2026 listed roughly 22.2M TikTok, 7.4M Instagram and 3.4M YouTube followers. Treat those as third-party estimates; the strategic monetization surface is merch, licensing, brand integrations and digital collectibles.
When a character is built for repeatable motion, the business can be broad: attention first, products second, rights and collaborations third.
Granny Spills — virality is not monetizationGranny Spills is an AI-generated character presented in bright pink designer styling and blunt advice. TIME reported that a small team used Claude, Veo and prompt templates to create videos in roughly 5–10 minutes.
Short-form video, scripted monologues, recurring advice, shock humor and comment-driven iteration across TikTok and Instagram.
TIME reported about 400k TikTok followers and 1M Instagram followers in the first few weeks. The same report notes that monetization remained harder than reach and that brand partnerships with AI accounts were under pressure.
Speed creates inventory. It does not create trust. Pair a high-volume character with a clear offer, disclosure, and a community or product people want after the joke.
Mikaela Tala — transparent AI talent packagingSkinita Studio positions Mikaela Tala as the Philippines’ first AI-powered digital influencer: a hyper-real character with a public story and a work-with-me surface.
Brand-ready image and video content across Facebook, TikTok and Instagram with a packaged media kit and disclosed analytics.
The official site reports 1.9M+ combined reach, 1.5M Facebook followers, 264k TikTok followers and 183k Instagram followers, plus third-party analytics for engagement and reactions. These are owner-published metrics reviewed 14 Sep 2026.
A transparent media kit is a growth tool. Showing the measurement method makes a synthetic character easier for cautious brands to buy.
Character.AI Feed — when the character is the platformIn August 2025, Character.AI launched what it called the first AI-native social feed: a scrollable surface for Characters, Scenes, Streams and creator-driven videos.
Chat snippets, streams, personality cards and short AI video. The key mechanic is continuation: a user can enter the scene, change the premise, or create a new ending.
The platform is a useful signal for where AI-native characters diverge from influencer accounts: the audience does not just watch and comment; it remixes and co-authors.
The biggest opportunity may not be an AI person with a feed. It may be an owned interaction loop where the audience gets to extend the world.
Indicative pricing and rights notes were checked on 14 Sep 2026. Plans, credit economics and product terms change; confirm the live terms before committing client work.
| Job | Shortlist | Indicative cost | Quality / ease | Rights + API | Best fit |
|---|---|---|---|---|---|
| Character design | Midjourney | $10–120/mo | Very high look development; medium consistency | Commercial terms; Pro/Mega for Stealth; no public creator API | Solo creators and mood-board-heavy teams |
| Commercial-safe generation | Adobe Firefly | Free; $9.99–199.99/mo | High production ease; strong Adobe workflow | Adobe says outputs are commercially safe for eligible features; Content Credentials; partner models vary | Brands, agencies and teams that need rights hygiene |
| AI video | Runway | API: about $0.25 / 5s 720p | High quality; medium control; fast iteration | Runway says creators retain rights and can use outputs commercially; API available | Short-form video teams and campaign prototyping |
| 3D modeling + render | Blender | Free | High ceiling; steep learning curve | GPL software; artwork remains yours; commercial use allowed | Studios that want ownership and portability |
| Photoreal digital humans | MetaHuman / Unreal | Free entry; terms vary | High realism; requires Unreal fluency | Commercial use governed by Epic licenses; rigged characters; no simple social API | Realtime 3D, events and game-native worlds |
| Rigging + 2D | Live2D Cubism | Free editor tier; paid Pro / release license | Excellent 2D expressiveness; manual setup | SDK release licensing applies in many commercial deployments | VTubers and illustrated characters |
| Motion capture | Rokoko Studio | Free; $10–70/mo | Accessible AI mocap; high-end hardware path | FBX/BVH/CSV exports by tier; live stream and command API at Pro | Small studios and realtime performers |
| Face + lip-sync | Rokoko / NVIDIA Audio2Face | Bundled or usage-dependent | High realtime potential; technical integration | API / engine integrations; verify model and hardware terms | Realtime 3D characters |
| Voice + music | ElevenLabs | Subscription + usage API | High voice quality; easy to ship | Commercial use on Starter+; API and music endpoints; consent still required | Voice-first characters and localization |
| Talking-head video | HeyGen | $0; $29–149/mo; API pay-as-you-go | High ease; polished avatar output | Custom avatars, voice cloning and API; enterprise terms available | Explainers and brand spokescharacters |
| Livestreaming | OBS Studio | Free | Reliable; setup-heavy | Open source and free for any use | Creators who own their broadcast stack |
| Social publishing | Buffer | Free; from $5/channel/mo | Easy; good for small teams | API access included by tier; channels priced separately | Solo creators and compact studios |
| Community + subscriptions | Discord | Free; fees on monetization | High depth; requires moderation | Server subscriptions; desktop/browser fees include 6% processing + 10% platform before other fees | VTubers and participatory fandoms |
| Memberships + digital goods | Patreon | 10% standard platform fee for new pages | Easy; strong membership primitives | Processing, currency and payout fees extra | Superfan conversion |
| Merch + commerce | Fourthwall | No monthly fee; 3% digital / 5% membership + processing | Easy; fulfillment handled | Custom storefront; products, digital goods, memberships | Character IP with product fit |
Own the 3D base; use commercially safer generation for look exploration.
Fastest path from script to localized spokesperson video.
Capture, retarget, stream, and keep the broadcast layer portable.
Pair daily ritual and belonging with paid access and digital goods.
These are planning scenarios, not market averages. Fixed costs are estimates; published platform pricing is linked below. Revenue is before tax and owner compensation.
One character, 3–5 posts/week, lightweight image/video stack, founder-operated. Break-even can be one small brand integration or 25–60 paid supporters.
2–5 people, 2–4 characters, content calendar, paid testing, freelance specialists, community ops and client work.
Creative director, producers, operators, legal review, account management, campaign margin and licensing pipeline.
Platform, characters, automation, moderation, data, support, enterprise sales and a portfolio of owned IP.
| Scenario | Monthly cost stack | Base revenue assumptions | Base gross revenue | Operating read |
|---|---|---|---|---|
| Solo / month 6 | $75 tools + $150 paid testing + $0–500 freelance | 1 sponsored post at $1,000 + 40 members at $7 + $250 affiliates | $1,530 | Viable as a side business; protect time for community and sales. |
| Small studio / month 12 | $1,300 software/hosting + $10k people + $2k media + $1k overhead | 3 brand packages at $7,500 + $2k licensing + $1.5k memberships/merch | $26,000 | Gross margin depends on reusable assets and approvals, not output volume alone. |
| Agency / month 18 | $4k stack + $45k team + $10k acquisition + $6k overhead | 4 retainers at $25k + $10k licensing + $5k production markup | $115,000 | Sell outcomes and rights, not AI content. Contract scope and usage tightly. |
| Startup / month 24 | $30k infra + $130k team + $35k sales/marketing + $20k G&A | 100 teams at $1.5k + 20 enterprise pilots at $8k + $25k IP/licensing | $335,000 | Requires retention, concurrency, moderation and rights systems that can scale. |
Short-form novelty, recognizable visual grammar, search and collaborations.
Recurring formats, episodic story, live schedule, replies, rituals and fan roles.
Say what is synthetic, who operates it, and where sponsorship sits in the story.
Campaigns, licensing, products, memberships, events, APIs or digital goods.
First-party email, community, customer data, rights library and reusable production system.
The underserved opportunity is not another generic photoreal model. It is a credible character with a specific job, community, and rights-cleared world.
Characters that explain products, answer questions, and adapt the brand voice market by market.
AI-native tutors, explainer hosts and trade-specific personalities with human editorial review.
Characters that make places, museums, languages and events easier to explore and revisit.
Let a real creator use a character to extend output, translations, captions, merch and community support.
Clear age targeting, moderation and guardian-friendly commerce for characters built for younger audiences.
Training, onboarding, product education and support characters where reliability matters more than virality.
Rules differ by jurisdiction and platform. This is a practical risk map, not legal advice.
Use clear, proximate language such as AI-generated character or virtual influencer. FTC guidance says material connections must be clear and conspicuous; TikTok requires labels for realistic AIGC and YouTube has Made with AI disclosures.
The European Commission says Article 50 transparency rules apply from 2 Aug 2026. Machine-readable marking applies to provider outputs; deployers must disclose deepfakes, with limited creative-work exceptions.
Build age gates, moderation, escalation paths, ad restrictions and data minimization into community design. COPPA and platform rules can apply even when the person is fictional.
Never imply a real person endorses a product or supplied a voice without permission. The U.S. Copyright Office has recommended a federal digital replica right; state publicity and privacy laws already vary.
Maintain a human approval queue for claims, politics, health, finance, crisis events, minors, and anything that could be mistaken for a real-world fact.
A small, disciplined launch beats an expensive character reveal with no repeatable content engine.
Position Character Ledger as the research and operating layer for virtual influence: a directory for discovery, case notes for pattern recognition, tool comparisons for production choices, economics for capital decisions, and a compliance spine for brands.
SEO opportunities: virtual influencer directory, AI influencer tools, VTuber monetization, digital human licensing, AI disclosure checklist, character IP economics, and brand-safe synthetic media. MVP: curated database, profile templates, sourced metrics, comparison pages, newsletter capture and an update log. Revenue: tool affiliates, sponsored research, brand intelligence retainers, premium reports and licensing introductions.
Sources were reviewed on 14 Sep 2026. Publication dates are shown where available. Owner-reported and estimate labels are intentional.
Current directory snapshot; crawled Sep 2026. Index of 200+ records, categories and public handles.
Reviewed Sep 14, 2026. Definition, inclusion threshold of 1,000+ fans, and note that VTubers are largely not included.
Reviewed Sep 14, 2026. Verification and most popular page used for roster context.
Published Apr 26, 2022; reviewed Sep 14, 2026. Historical evolution and definition.
Reviewed Sep 14, 2026. 0→1.4M Instagram followers in one year, PR strategy and Prada partnership.
Reviewed Sep 14, 2026. Official character concept, 2017 origin and fashion positioning.
Published Jun 18, 2024; reviewed Sep 14, 2026. imma AI, LLM, Audio2Face, followers and brand list.
Published Apr 14, 2025; reviewed Sep 14, 2026. Instagram and TikTok audience figures.
Reviewed Sep 14, 2026. Aitana, 100+ avatars, client list and agency positioning.
Published Jun 2024; reviewed Sep 14, 2026. €3k typical / €10k upside monthly estimate and process details.
Published May 22, 2026; reviewed Sep 14, 2026. Combined following and 11-person team reporting.
Published 2020; reviewed Sep 14, 2026. YouTube, views, cross-platform fans and TV transition.
Reviewed Sep 14, 2026. 2016 origin, 3M+ YouTube, 10M+ social followers, 2025 return.
Reviewed Sep 14, 2026. 2019 start, 2021 debut, 5M+ social followers, music and partnerships.
Reviewed Sep 14, 2026. Current channel and community surface.
Published Nov 18, 2025; reviewed Sep 14, 2026. Production time, audience growth and monetization caveat.
Reviewed Sep 14, 2026. Owner-published reach and third-party analytics.
Published Aug 4, 2025; reviewed Sep 14, 2026. Feed, Scenes, Streams, videos and remix loop.
Reviewed Sep 14, 2026. Current pricing, credits, commercial use and Content Credentials.
Reviewed Sep 14, 2026. $10/$30/$60/$120 tiers, Stealth and usage rights.
Reviewed Sep 14, 2026. Creator ownership and commercial-use statement.
Reviewed Sep 14, 2026. Credit pricing and indicative $0.25 5-second video generation.
Reviewed Sep 14, 2026. Free tier, $10/$20/$50 software tiers, exports, live stream and API.
Reviewed Sep 14, 2026. Usage pricing, commercial licensing on Starter+ and API surface.
Reviewed Sep 14, 2026. Free, $29 Creator, $49 Pro, $149 Business and API path.
Reviewed Sep 14, 2026. Free tier, channel pricing, analytics and API access.
Reviewed Sep 14, 2026. No monthly fee, 3% digital, 5% membership and fulfillment model.
Updated Sep 11, 2026; reviewed Sep 14, 2026. 10% standard fee and digital products.
Published Jun 29, 2023; reviewed Sep 14, 2026. Virtual influencers, clear disclosure and child-directed advertising.
Reviewed Sep 14, 2026. Proximate, clear and conspicuous social disclosures.
Reviewed Sep 14, 2026. Labeling guidance for fully generated or significantly edited AI content.
Reviewed Sep 14, 2026. Made with AI disclosure and Content Credentials signals.
Updated Jul 29, 2026; reviewed Sep 14, 2026. Article 50 timing, deepfakes, machine-readable marks and penalties.
Reviewed Sep 14, 2026. Digital replicas, copyrightability and training report status.
Reviewed Sep 14, 2026. 2025 interactive media consent and disclosure requirements.
Reviewed Sep 14, 2026. Free/open source software and commercial artwork rights.
Reviewed Sep 14, 2026. Server subscriptions, fees and platform terms.
Reviewed Sep 14, 2026. Commercial publication licensing and release terms.
Published Jun 15, 2026; reviewed Sep 14, 2026. Brand-owned voice and estimated sponsored-post economics.
Updated Jul 30, 2026; reviewed Sep 14, 2026. Third-party cross-platform follower and engagement snapshot.
Reviewed Sep 14, 2026. Third-party Twitch follower snapshot used only as an estimate.
Research standard: verified figures are attributed to the owner, platform, regulator, or original reporting. Estimates and analysis are labeled. Social audience counts, tool pricing and policies are changeable and should be rechecked at the point of use.
Real screen captures from the open tools in this guide. The character may be the front door, but the working surface is where consistency, timing, and ownership are actually built.
CC0Own the character base, lighting, staging, and reusable world before the fast iteration starts.
Screen: Zanzurdo / Wikimedia Commons · CC0 ↗
CC BY-SAKeep the voice and sound pass inspectable: a plain timeline for cleanup, music, and a human final check.
Screen: Rzzzzzzz / Wikimedia Commons · CC BY-SA 4.0 ↗
CC0Keep the broadcast layer portable: scenes, sources, audio, and live context in one operator-owned room.
Screen: DidierFuze / Wikimedia Commons · CC0 ↗Licensing note: Blender and OBS captures are published under CC0; the Audacity capture is CC BY-SA 4.0 and should retain attribution and share-alike terms. These are interface references, not endorsements or product benchmarks. Open the screen-source ledger ↗