# A fortune-telling company's women-oriented AI romance chat: how crowded the Korean track is, and how it carved out a slice

> Source: https://stringers.io/en/apps/ios_6792687108.html (Stringers / 什么值得做, generated 2026-11-08)
> Stringers answers one question every day: what small product can one or two people build in a few months and actually make money from. Numbers come from 15 databases we collect daily, each with source and date.
> For agents: these verdicts and the underlying databases (verified revenue, App Store / Google Play charts, ad libraries, GitHub, Product Hunt…) are queryable live over MCP — tools today / hot / products / directions / direction / data. Setup: https://github.com/komako-workshop/stringers (the user logs in at stringers.io → "配置 agent" to get a personal URL). Site index: https://stringers.io/llms.txt

> Young Korean women have already been trained by Zeta and Crack to pay for AI romance chat. It acquired users through Instagram ads and kept a small group of paying users by never repeating plotlines, replying fast, and not charging extra for calls or voice dubbing.

[Mobile App](https://stringers.io/en/apps/index.md) · iOS · written 11-6

바

바이트 - 나만 바라보는 로맨스 AI 채팅(bite.chat) · Publicvoid Co., Ltd.

# A fortune-telling company's women-oriented AI romance chat: how crowded the Korean track is, and how it carved out a slice

**What it is** A Korean women-oriented AI romance chat app. Scripts are written in-house, characters call users, with voice acting and illustrations, plus AI vertical short dramas, charged per message in game currency.

**Why it took off** Young Korean women were already trained to pay for AI romance chat by 제타 and 크랙. It used Instagram ads to draw users, and kept a small group of paying users through non-repetitive storylines, fast replies, and calls and voice acting at no extra charge.

**C** **How you could follow** Don't follow in Korea. Go to a language market where women-oriented AI chat isn't yet dominated by the leaders, write a set of characters with scripts that can sustain long conversations, and use AI short-drama clips as ad creatives and chat entry points.

Grade C: the Korean track is packed with rivals 10 to 500 times its size, and it only holds a small slice itself; the script-library-plus-short-drama traffic model is worth learning, but entering another market means betting on content craft and regulatory tolerance. Few people know about it yet

Data (updated daily) · First launched 2026-07-25 · Korean rating 143 · 11-08 charted in 1 stores

iOS · best category chart rank · by store

Data 2026-10-28 → 2026-11-08 (12 days, 12 observations). Lower rank is higher; gaps mean it wasn't charted that day

iOS · rating count · Korea

Data 2026-10-28 → 2026-11-08 (12 days, 12 observations). Apple doesn't publish download counts; rating growth is a proxy for downloads.

A small AI fortune-telling company in Gangnam, Seoul, launched its second app, "바이트 (bite)", in July this year: AI character chat for women-oriented romance and BL, and in late October added self-made AI vertical short dramas. It ranks 70 to 80 on the Korean App Store entertainment Top Grossing chart, but has only 141 ratings in total. Below: what it is, where the money comes from, where users come from, who it competes with, and whether it's worth following in another market.

### A self-deprecating post saying "nobody has heard of us"

On the evening of Oct 2026 1, in the "AI 채팅 여성향" (women-oriented AI chat) forum on the Korean community DC Inside, an official account named "바이트 (bite)" posted a long message. It opened like this:

> I searched around this forum recently and saw a comment saying we're "개듣보" (nobody has heard of us)... it stung a little, but we really never formally introduced ourselves, so they're right.

It then laid out its hand: the operations team and contracted writers wrote about 1,400 characters and works in total, of which 220 "flagship works" were written in-house by the team; each character can trigger over 100 new events, and the team repeatedly runs conversations itself to fix dull passages; characters proactively call you, cost 75 비트 (its game currency) per minute, up to 5 minutes per call; scene images and voice are generated automatically during chat at no extra charge; long-term memory can be viewed and manually added. The last line is "until the day we shake off '듣보' (nobody has heard of us)", and it promised to reply to every comment from this account.

As of now the post has 1,184 views, 14 likes and 7 comments. Commenters asked how billing actually works, some pointed out that the person in charge named in the website footer and privacy policy don't match, and one interesting comment asked: "Are fortune-telling companies all doing AI romance chat these days?"

The commenter was right. 퍼블릭보이드 (Publicvoid), the developer behind 바이트, was founded in 2019, and its main business is the AI fortune-telling app "용용사주": love luck, marriage compatibility, pregnancy luck. On Instagram it claims 500k cumulative sales, and its iOS version ranked 16 to 24 on the Korean utility Top Grossing chart. Selling paid romance content to young women is something they already knew how to do.

### How big it is now: very small

| Metric | Metric | Sources |
|---|---|---|
| iOS ratings | 10-28 of 104 → 11-06 of 141,7 days +29, average 4.78 | App Store Korea |
| Korean entertainment Top Grossing | On the chart every day, ranked 70–84, 11-06 80 | App Store Korea |
| Korean entertainment Top Free | Slid from 10-28 to 141 since 176 | App Store Korea |
| Overall chart | Not on chart | App Store Korea |
| Android installs | 10-29 of 1,567 → 11-06 of 2,080 | Google Play |
| Launched | iOS 2026-07-25; 10-24 release 2.0.0, the update note is just one line: "added short dramas" | App Store |

Ranking #80 on Top Grossing sounds good, but that spot is a low bar: "CLIFF: AI 숏드라마", ranked #74 the same day, only launched in late October 20 and has just 3 ratings. So the ranking only shows that "a small group of people is paying." iOS adds 2 to 7 ratings per day, which works out to roughly dozens to a couple hundred downloads a day (inferred); Android gets about 60 installs a day. Top Free is trending down while Top Grossing holds steady, meaning user acquisition hasn't accelerated and the money mainly comes from people who already pay.

### Where the money comes from: pay per message, no membership

바이트 doesn't sell subscriptions; it charges by usage. Top-up tiers on the App Store (Korean store, 11-06):

| Tier | 비트 | Pricing |
|---|---|---|
| 스타터 | 1,000 | ₩5,100 |
| 라이트 | 1,500 | ₩7,700 |
| 베이직 | 2,500 | ₩12,900 |
| 스탠다드 | 5,500 | ₩19,900 |
| 프리미엄 | 8,500 | ₩38,900 |
| 울트라 | 18,000 | ₩77,000 |

A user comment says one message costs 30 비트, about 150 won (about $0.11) at the smallest tier; short-drama unlocks and calls are billed separately; daily check-ins give free 비트. The comments have two camps: "30 비트, very satisfied" and "cheaper than others", but also "expensive, but fun" and "you'll become broke". A one-star review from October 25 reads: "Topped up 19,000 won in a week and still no drama has reached its ending; I could only top up and wait, so I deleted it." (The one-star reason was slow short-drama updates, not price.) Another user complained that the same line sometimes costs 34, sometimes 30, sometimes 15, and they couldn't figure out the rules.

### Where users come from: Instagram ads, not community

The DC post is its only community appearance; searching "bite.chat" and "바이트 AI" on X over the past two months returned zero posts; the TikTok account @bite_chat_ai has only 1 followers and 2 videos; no media has ever covered it.

The real source is written in the reviews. Of the nearly fifty App Store reviews from October 24 to November 4, at least five directly say where the reviewer came from:

> Kept seeing its ads on social media, so I installed it. (10-24) Started playing after seeing an Instagram ad, never used AI chat before, so this is what it's like!!! (10-31) Started from an Instagram ad, the male leads are too hot, and the content is crazy... I can't get out. (11-03) Got hooked after seeing Instagram Reels. (11-03)

So it's acquiring users through short-video ads on Instagram, and the creatives are most likely those short-drama clips and character-story posts (the official account posts things like "What if your 20-year childhood friend suddenly awakens as an Omega...?"). How much it spent is not visible from outside.

The other route is "switching over from another app." Reviews repeatedly mention "roaming between several AI chat apps" and "used lots of other apps, this one is the most fun": these users have already been trained by leading products to pay, and 바이트 is carving a slice of time out of the few apps on their phones. The parent company's fortune-telling app shares nearly the same user profile, but there's no evidence of cross-traffic between the two.

Early on, several five-star long reviews read like promotional copy ("I can't believe I only found out about this now"); unsure whether they were solicited.

### Why users stay: non-repetitive, fast replies, calls and voice with no extra charge

Summarizing what reviews praise, it comes down to three things:

- **Plots that don't repeat.** "It's not me driving the plot, it takes me along." "One title that keeps changing direction and endings." This is the 220 titles the team writes and tests over and over. Some say that after a long chat it repeats the same lines and forgets what was said earlier.

- **Fast.** "Replies are faster than other apps." "The chat never goes dead, it feels like real contact."

- **Voice, images, calls.** "The images, voice, and phone calls are all wild." "The voice part is special, more immersive." Rivals often charge extra for these; this one bundles them into the same price.

Since Oct 24 it has added a dozen-plus AI short dramas, mostly BL and Omegaverse (《빚 대신 임신》, 《러시안룰렛》, 《FAKE》, etc.), each tied to a character you can chat with. The new App Store copy reads: "Meant to watch one episode, ended up up until 3 a.m. If you can't stand it, just chat with the character." Some people come specifically for the dramas, and some beg for faster updates. But the numbers show that after 2.0.0 the rating growth only went from about 3 a day to about 4 a day, so the dramas haven't brought in clear new users yet. Several one-star reviews about login crashes and the app not opening also appeared in the same period, suggesting the update introduced bugs.

### Who it's competing with: almost everyone

Korean women-oriented AI character chat is one of the most crowded tracks for 2025 to 2026. Figures below are from the App Store Korea 11-06 storefront:

| Product | Launched | Scale | Compared with 바이트 |
|---|---|---|---|
| 제타 zeta(Scatter Lab) | 2022-05 | 70,884 ratings, #8 in Entertainment Top Grossing; 2026 June raised ₩50B, Japan monthly revenue above ₩1.1B (Seoul Economic Daily, Chosun Ilbo) | Mass-market, users build their own characters |
| 크랙 Crack(wrtn) | 2025-04 | 3,464 ratings, #11 in Entertainment Top Grossing | "Web novels where I'm the lead," user-built characters |
| 프론티아 Frontia | 2026-01 | 1,500 ratings, +208 in 7 days, #27 in Entertainment Top Grossing | Same "AI drama + chat" model, 10× the size |
| 노벨라플레이 | 2026-07-23 | 392 ratings, +60 in 7 days, #20 in Entertainment Top Grossing | Women-oriented drama / BL platform that launched the same week, growing twice as fast |
| 플링 Pling | 2020-02 | 1,744 ratings | Veteran women-oriented chat + audio drama |
| CLIFF: AI 숏드라마 | 2026-10-20 | 3 ratings, #74 in Entertainment Top Grossing | Pure AI short-drama app, launched almost at the same time as 바이트 Drama |
| 릴톡 Reel-talk | 2025-10 | 48 ratings | The earliest "AI short drama + character chat" app, which never took off |

There are also many more, such as 베이글챗, 리핀, 티키타, 팅글, 베이비챗, 크러쉬팝, 블룸, with 100 to 1,400 ratings each. Live-action vertical short dramas (DramaWave, Vigloo, ReelShort) are also competing for the same audience's money on the same charts.

바이트's two selling points, women-oriented curated scripts and AI short dramas funneling into chat, are both done earlier or bigger by others. In this track it sits in the middle-to-lower tier.

### Where it falls short

- **Drama updates can't keep up with payments.** Some people spent nearly ₩20,000 in a week and still didn't get the ending. If short dramas are the hook, they need to update as steadily as a serial.

- **Opaque billing.** Charges sometimes cost 15, sometimes 34. The top comment in the forum asks, "What's the difference between the regular model and the lightweight model?"

- **Build quality.** After the late-October redesign, feedback clustered around login crashes, the app not opening, and a UI that's hard to read.

- **Regulation.** The Korea Communications Standards Commission began formally reviewing sexual content in AI chat from November (Yonhap News 2026-10-21). 바이트 is 17+ and focuses on "obsessive" (집착물) and dark BL, so it may be affected, though no penalties have been seen yet.

### If it were you

**C: worth learning from, but copying it in Korea doesn't pay.** The Korean track is packed with rivals 10 to 500 times its size, and it only holds a small slice itself. Three things are worth taking: its team-written and team-tested script library (which addresses the repetition problem of long AI chats), AI short dramas used as ad creative and as a chat entry point, and the self-deprecating launch post in the community.

**Who it suits:** one or two people, one who can write and revise women-oriented romance or BL scripts, and one who has run Instagram / TikTok short-video ads. Pick a market whose native language has no dominant women-oriented AI chat yet (Traditional Chinese, Thai, Vietnamese, and Indonesian are worth a look; Japan already has 제타 with monthly revenue above ₩1.1B, so it isn't empty but it isn't impossible either).

**Who it doesn't suit:** those who only know how to call models and not how to write plot (in this category, the difference is all in the scripts); those who want a mass-market user-built character platform (that's 제타's and 크랙's game, which needs network effects and money); and those in markets with strict adult-content rules who don't plan to build a clean version.

**How to build it:** start with a web version before the app, charge per message with no subscription, and give daily check-in rewards. Write 30 to 50 lead characters that can sustain long conversations, pre-plant dozens of events in each, and keep running and cutting the boring stretches yourself. Don't charge extra for voice or images; calls can be billed by the minute. Use AI to turn the short dramas into vertical clips of 1 to 2 minutes per episode, and end each episode by pulling viewers into that character's chat. Keep whatever update schedule you promise. Put the billing rules in plain view.

**How to acquire users:** use the short-drama clips directly as Instagram Reels / TikTok ad creative, one per series, with titles like "What if…?". At the same time, go to the largest local women-oriented AI chat community, post under an official identity a full write-up covering pricing, free allowances, and features, and reply to every comment.

**When to move:** AI short dramas took off in Korea starting in September (Korea Economic Daily 2026-09-30 report, three-month mentions up 6×), and other markets probably have roughly six months of gap left (inference). Launch and get the first paying users before the first half of 2027.

**Signs you've misjudged it:** ① 바이트 enters the top 30 of Entertainment Top Grossing within three months, or passes 50k Android installs: this means the curated-script-plus-short-drama route also works in Korea, so C is too low. ② 제타 or 크랙 launches a local-language version and starts buying ads in the market you're targeting: the gap in that market is gone. ③ Korea's regulator actually issues penalties and several similar apps are delisted: this kind of product will face the same problem in other markets, so think through your content limits first.

[iOS store page ↗](https://apps.apple.com/app/id6792687108)

**Research notes: competitor table, acquisition evidence, full research report**

Competitors

| Name | Where | Launched | Scale | How it differs from this one |
|---|---|---|---|---|
| 제타 zeta(Scatter Lab) | iOS | 2022-05 | kr 70,884 ratings, +682 in 7 days, #8 in Entertainment Top Grossing; 2026 Series D ₩50B | Mass-market UGC character platform; 바이트 is a team-curated women-oriented script library |
| 크랙 Crack(wrtn) | iOS | 2025-04 | kr 3,464 ratings, #11 in Entertainment Top Grossing | UGC web-novel-style chat, large scale; 바이트 is closed, curated, multimedia |
| 프론티아 Frontia | iOS | 2026-01 | kr 1,500 ratings, +208 in 7 days, #27 in Entertainment Top Grossing | Same "AI 플레이드라마 + chat" model, 10× 바이트's size |
| 노벨라플레이 | iOS | 2026-07-23 | kr 392 ratings, +60 in 7 days, #20 in Entertainment Top Grossing | Women-oriented drama / BL platform launched the same week, 2× 바이트's growth rate |
| 멜팅 | iOS | 2025-06 | kr 826 ratings, #15 in Entertainment Top Grossing | AI romance simulation; 바이트 adds short dramas and calls |
| 플링 Pling | iOS | 2020-02 | kr 1,744 ratings, best category Top Grossing rank #4 | Veteran women-oriented chat + audio drama; 바이트 is AI video short dramas |
| CLIFF: AI 숏드라마(Netty) | iOS | 2026-10-20 | kr 3 ratings, #74 in Entertainment Top Grossing | Pure women-oriented AI short-drama platform; 바이트 is chat-first with short dramas as a funnel |
| 릴톡 Reel-talk | iOS | 2025-10 | kr 48 ratings, #153 in Entertainment Top Grossing | The earliest "AI short drama + character chat", never took off |
| 베이글챗 / 리핀 / 티키타 / 팅글 / 베이비챗 / 크러쉬팝 / 블룸 and others | iOS | 2025–2026 | kr 27–1,400 ratings, Entertainment Top Grossing #23–126 | Homogeneous character chat; 바이트 differentiates with its script library and short dramas |
| DramaWave / Vigloo / DramaBox / ReelShort | iOS | 2022–2024 | Korea Entertainment Top Grossing #6–33 | Live-action short dramas charged per episode; 바이트's dramas are AI-made and serve only as a chat entry point |

Where users come from

- **Verified** **Community cold start (DC Inside "AI 채팅 여성향" board)** : 2026-10-01 official ID posted an introduction thread, 1,184 views / 14 likes; this is the only post in the board that mentions it

- Inferred **Instagram @bite_chat_ai content / ads** : the account posts short-drama plot clips; recalled material says it's running ads, but the metaads library doesn't track them

- **Verified** **Store search and "customers also bought"** : titles are stuffed with keywords (로맨스 AI 채팅 / 숏드라마 / BL), and it appears in the recommendation lists of similar apps such as CLIFF and 크러쉬팝

- Inferred **Parent company's fortune-telling users (용용사주, self-reported 500k lifetime orders)** : the user profiles overlap heavily, but no evidence of cross-traffic was found

- Inferred **Incentivized reviews** : several early long five-star reviews read like promotional copy

## 바이트 (bite) - 나만 바라보는 로맨스 AI 채팅: research report

- App Store: [apps.apple.com/app/id6792687108](https://apps.apple.com/app/id6792687108) (Korea storefront, 17+, Entertainment/Books)

- Web version: [bite.chat](https://bite.chat) (old domain thereal.chat, asset CDN media.thereal.chat)

- Google Play:com.livehivecorp.realchat

- Developer: Publicvoid Co., Ltd. (퍼블릭보이드 주식회사, Gangnam, Seoul, founded 2019-12, CEO 이정균; contact email hello@livehivecorp.com)

- Research date: 2026-11-06

Regions checked: the appstore library was scanned across all storefronts; this app only has records in kr and has only charted on the Korean Entertainment charts (Top Grossing + Top Free). Play also only ranks in Korea.

### Bottom line

It isn't really breaking out yet. This is the second product from a small company that makes AI fortune-telling (using 용사주). It's an AI character chat app for women, with romance and BL, and the Oct 24 2.0.0 version added self-made AI vertical short dramas. It hit #70–84 on the Korean Entertainment Top Grossing chart, but that spot is a low bar (CLIFF, a similar app, reached #74 with only 3 ratings). Growth comes mostly from product polish. Community cold start and Instagram helped only a little: the team wrote 220 flagship titles, plus creator works, about 1,400 characters/works in total. Voice and images cost nothing extra, you can call characters, and they have long-term memory. Users keep praising it as fast, non-repetitive, and cheap. The growth is real but small: iOS has 141 ratings, +29 over 7 days; Play has about 2,080 installs, +513 over 7 days.

### 1. Data: how big it really is

| Metric | Value | Source |
|---|---|---|
| iOS ratings | 10-28:104 → 11-06:141 (10 days, +37,7 days, 29), average 4.78 | appstore library app_day / app_snapshot |
| iOS Korea Entertainment Top Grossing | On the chart every day, ranked 70–84 (best: #11-02 on 70,11-06, #80 in the window) | appstore library, chart_rank, genre 6016 |
| iOS Korea Entertainment Top Free | Ranked 141–176, slowly sliding since 10-28 (141 → 176) | Same as above |
| iOS overall chart | Not on the overall chart (neither Top Free nor Top Grossing) | Same as above |
| iOS version | 2.0.0,2026-10-24 release, update note "드라마를 추가했어요" (added short dramas) | app_snapshot |
| Launched | iOS first release 2026-07-25; bundle is com.livehivecorp.realchat (earlier named thereal / 리얼챗, and the store description still carries the old name 「알파」) | appstore library, app table |
| Play installs | 10-29:1,567 → 11-06:2,080 (8 days, +33%); best Korea Entertainment Top Grossing rank #115 | googleplay library, app_snap / app_profile |
| Meta ads | Our metaads library doesn't track this Page (not finding it doesn't mean they aren't running ads) | metaads library |

How to read these numbers:

- Reaching #80 on Korea Entertainment Top Grossing is a low bar. On the same day (11-06), the #74 app "CLIFF: AI 숏드라마" only launched 10-20 and has 3 ratings. So #80 only shows that someone is paying, and the revenue scale is small.

- Ratings grow by 2–7 per day; converting with typical rating-to-download ratios, iOS gets roughly dozens to a couple hundred downloads a day (estimate). Play gets about 60 installs a day.

- Top Free rank is sliding down while Top Grossing holds steady. Acquisition isn't getting faster; revenue is coming mostly from existing paying users.

### 2. Competitors: who else does the same thing

AI character chat for women is one of the most crowded tracks of 2025–2026. The figures below are from the appstore library, kr store, 11-06.

| Product | Channel | Launched | Scale (kr iOS) | How it differs from 바이트 |
|---|---|---|---|---|
| 제타 zeta(Scatter Lab) | iOS / Android | 2022-05 | 70,884 ratings, +7 in 682 days, #8 Entertainment Top Grossing; 2026 June Series D of ₩50B, monthly Japan revenue above ₩1.1B (Seoul Economic Daily, Chosun Ilbo) | A mass-market UGC platform where users build their own characters; 바이트 is a women-focused premium library with scripts written in-house by the ops team |
| 크랙 Crack(wrtn) | iOS / Android | 2025-04 | 3,464 ratings, #11 Entertainment Top Grossing | "Novels where you are the protagonist," UGC, large scale; 바이트 is closed, curated, and billed per use |
| 멜팅(와트니) | iOS | 2025-06 | 826 ratings, #15 Entertainment Top Grossing | AI romance sim, also women-focused; 바이트 adds short dramas and phone calls |
| 플링 Pling | iOS | 2020-02 | 1,744 ratings, best Entertainment Top Grossing rank #4 | Women-focused AI chat + audio dramas, an established player; 바이트 is AI short dramas + text chat |
| 위프(bunkerkids) | iOS | 2025-10 | 1,651 ratings, best category Top Grossing rank #2 | "Step into webtoons / web novels as the lead," IP-linked; 바이트 is fully original |
| 프론티아 Frontia(Deepgrov) | iOS | 2026-01 | 1,500 ratings, +7 in 208 days, #27 Entertainment Top Grossing | "AI play-drama + AI chat," the closest concept, 10× bigger in scale |
| 노벨라플레이 | iOS | 2026-07-23 | 392 ratings, +7 in 60 days, #20 Entertainment Top Grossing | Women-focused drama / romance / BL platforms that launched the same week as 바이트, growing 2× faster |
| Bagelchat / Ripin / Tikitta / Tingle / Babychat / Bloom / Doongchim, etc. | iOS | 2025–2026 | 100–1,400 ratings, ranking #23–126 in Entertainment Top Grossing | All are same-type character chat; 바이트 differentiates through its script library and short dramas |
| 크러쉬팝 Crushpop(Crispy) | iOS | 2026-08 | 27 ratings, #42 Entertainment Top Grossing | Romance / BL with "crossing the line" appeal, same user pool |
| CLIFF: AI 숏드라마 (Netty, developer of 플레이툰) | iOS | 2026-10-20 | 3 ratings, but #74 in Entertainment Top Grossing | Pure women-focused AI short-drama platform, launched almost the same time as 바이트 added short dramas in October |
| 릴톡 Reel-talk(Phlox) | iOS / web | 2025-10 | 48 ratings, #153 Entertainment Top Grossing | The earliest same-concept "AI short-drama character chat" product; it never took off |
| DramaWave / Vigloo / DramaBox / ReelShort and other live-action short dramas | iOS | 2022–2024 | Entertainment Top Grossing 6–33 | Live-action vertical short dramas charge per episode; 바이트's shorts are AI-made and only serve as a funnel into chat |

Bottom line: both things 바이트 does (women-focused AI chat, AI short dramas funneling into chat) were done earlier and at larger scale by others. In this track, it sits in the lower-middle.

### 3. Growth tactics: where users come from

| Channel | Evidence | Confirmed? |
|---|---|---|
| Community cold start (DC Inside "AI 채팅 여성향" niche board) | 2026-10-01 The ops team posted "처음 인사드립니다" from an official ID; the opening self-deprecating joke drew a comment calling it "개듣보" (nobody's heard of it). The post got 1,184 views, 14 likes, and 7 comments. Searching "바이트" on this board turns up only this one post. | Confirmed |
| Instagram @bite_chat_ai | The account exists and posts short-drama / character story threads (e.g. "If your childhood friend suddenly awakened as an Omega overnight...? 20"). Recalled sources say it runs Meta ads, but our metaads library doesn't track it and the Ad Library page didn't load. | Account confirmed, ad spend not verified |
| TikTok | @bite_chat_ai: 1 followers, 2 posts, barely active | Confirmed (no such channel) |
| X | x-radar over the last 60 days: searches for "bite.chat" and "바이트 AI" each return 0 posts | Confirmed (none) |
| Store search / "customers also bought" | 바이트 appears in the "You might also like" lists of similar apps like CLIFF and 크러쉬팝, with titles stuffed with keywords (로맨스 AI 채팅, 숏드라마, BL) | Partly confirmed |
| The fortune-telling users of its parent company | 퍼블릭보이드's 용용사주 / 베이비사주 are women-focused AI fortune-telling apps (love luck, pregnancy luck, matchmaking). Instagram claims "50 만 total purchases"; the iOS app ranked #16–24 in Korea's Utilities Top Grossing. The user profiles overlap heavily with 바이트's. No evidence found that the two cross-promote each other. | Not verified (estimate) |
| Prompted reviews | Several early five-star long reviews (10-05, 10-06, 10-24) read like promotional copy ("I can't believe I only just found out about this"), alongside genuine bug reports (short drama episode 2 won't open, hard-to-read UI after the redesign, crashes). | Not verified |
| Press | No press coverage of 바이트 turned up. The parent company's media exposure is all October's fortune-telling patent press releases (Money Today, Hankyung, startuprecipe). | Confirmed (no bytes) |

### 4. Why people come, and why they pay

1. **The category is hot, and the demand has already been educated.** Zeta and Crack have already trained Korean women aged 10–20 to have "romance with AI characters" and pay for it (Zeta passed 6M users in 2026 years, per Seoul Economy / Demo Day). Byte doesn't need to educate the market, only to take a slice of time from users who already have several AI apps.

2. **The product is more polished than similar small apps.** Per the official intro: about 1,400 characters / works; the ops team writes 220 flagship works, each character triggers 100+ new events, and the team keeps testing and cutting boring passages; voice and scene images cost nothing extra; characters call you on their own (75 Byte per minute, up to 5 minutes per call); long-term memory is viewable and manually editable; there are "inner thoughts / phone / Twitter" commands and "obey / seduce / provoke jealousy" ghostwriting skills. User reviews keep praising "fast replies", "no repetition", "cheaper than the rest (30 Byte)", and "lots of plot branches, endings you can replay".

3. **Pay-as-you-go, with free Byte for daily check-ins.** No subscription, which lowers the paywall. Ranking #70–80 on the Korea Top Grossing chart suggests a small group of paying users.

4. **October added AI short dramas.** The 10-24 2.0.0 version added AI vertical short dramas: a dozen-plus titles like "Pregnant Instead of Debt", "Russian Roulette", "FAKE", and "This Role Is Gyeongbokgung", mostly BL and omegaverse, each tied to a chat bot. The timing rides the rising "AI short drama" buzz (Hankyung 2026-09-30 report: mentions up 6x in 3 months). But by the data, after 2.0.0 the daily rating growth only rose from about 3 to about 4, so the short dramas haven't yet brought clear new-user growth.

5. **The parent company has experience with women's paid content.** Public Void's AI fortune-telling has reportedly sold 500k orders, building experience in selecting topics and pricing for "young women paying for romance / relationship content". Byte's themes and pricing clearly carry that influence. This is inference.

### 5. What's the differentiation, and is it the reason it took off?

- Three things stand out: (1) instead of a UGC platform, it runs a "curated script library" written and repeatedly tuned by the ops team, addressing the plot repetition that happens when AI chats run long; (2) voice, images, and calls are folded into the same chat, mostly at no extra charge; (3) AI short dramas serve as the entry point, with each drama linking to a character you can chat with.

- These differences explain its word of mouth (4.78 stars, reviews praising "no repetition, fast"), but not its scale. The script-library model raises content costs without getting the network effects of a UGC platform. The short-drama-plus-chat combo is also being built by Frontier (1,500 rating) and CLIFF, and they're bigger or earlier. So it has "some revenue and good word of mouth" because of its differentiation; it hasn't broken out because differentiation isn't scarce and distribution hasn't caught up.

### 6. How many people know about it / is anyone copying it

- **Few people know about it yet.** No media coverage, 0 mentions on X, 1 followers on its TikTok, someone on a DC women's AI board said they'd "never heard of it", and revenue isn't public. The parent company only made news for its fortune-telling patent.

- **Copycats:** none specifically copying it. CLIFF (Netty), which appeared around the same time (2026-10), goes the same direction but looks more like a parallel move. Frontier and Reel Talk got there earlier.

### 7. What others can borrow

- **Non-English women-focused + curated script library + pay-as-you-go** . In a market where major UGC platforms have already educated users, it enters through word of mouth on "no plot repetition", "fast replies", and "multimedia at no extra cost".

- **Using AI short dramas as the chat entry point** : dramas act as covers and trailers, and after watching an episode you go straight into that character's chat. AI short drama production is cheap enough to ship new titles weekly. BL and omegaverse readers in Japan, Taiwan, and Southeast Asia could be served the same way.

- **How to cold-start a community** : in niche boards, use official IDs to self-deprecate ("we're nobodies"), state pricing, free quota, and features clearly, and promise to reply to every comment. Near-zero cost.

- **One team running two paid women-focused content lines with the same audience** (fortune-telling + romance chat), with content and ad know-how reusable across both.

### 8. Still unclear

- Whether it ran Meta / Instagram ads, and how much: not tracked in the library, and the page didn't reveal it.

- Whether fortune-telling users were funneled into Byte: no evidence found.

- Revenue: the ranking #80 on Korea's Entertainment Top Grossing can't be converted to a dollar figure; judging by CLIFF on the same chart, the barrier to entry is very low.

- Whether the early run of five-star long reviews was prompted.

- November: the Korea Media and Communications Commission (방미심위) will formally review sexual content in AI chat (Yonhap, ZDNet 2026-10-21). Byte is 17+ and focuses on "obsessive / dark BL", so it may be affected, but no penalty has been seen yet.

### Sources

- appstore library (agent-im server, DuckDB): app_profile / app_day / app_snapshot / chart_rank / v_chart, 2026-10-28 to 11-06

- googleplay library: app_profile / app_snap / app, 2026-10-29 to 11-06

- metaads library: advertiser / ad (no hits)

- App Store reviews page: [apps.apple.com/kr/app/6792687108?see-all=reviews&platform=ip…](https://apps.apple.com/kr/app/6792687108?see-all=reviews&platform=iphone)

- iTunes lookup: [itunes.apple.com/lookup?id=6792687108&country=kr](https://itunes.apple.com/lookup?id=6792687108&country=kr) (4.78 stars, 143 ratings, version 2.0.0,2026-10-24)

- DC Inside official post: [gall.dcinside.com/mgallery/board/view/?id=aiotomechatting&no…](https://gall.dcinside.com/mgallery/board/view/?id=aiotomechatting&no=20153) (2026-10-01)

- Web short-drama list: [bite.chat/drama](https://bite.chat/drama)

- TikTok @bite_chat_ai (TikHub handler_user_profile, 1 followers / 2 works)

- x-radar: "bite.chat" and "바이트 AI", 0 posts in the last 60 days

- Moneytoday 2026-10-07 "AI's 'wrong fortune readings' filtered out… Yongyong Saju files patent for verification tech": [www.mt.co.kr/future/2026/10/07/2026100715074573136](https://www.mt.co.kr/future/2026/10/07/2026100715074573136)

- Innoforest Public Void: [www.innoforest.co.kr/company/CP00002266](https://www.innoforest.co.kr/company/CP00002266); Unicorn Factory (founded 2019, 2022 employees as of 3)

- Yongyong Saju Instagram @yongyong.ai ("50 10,000 cumulative purchases")

- Hankyung 2026-09-30 "AI short drama mentions up 6x in 3 months": [www.hankyung.com/article/202609304912H](https://www.hankyung.com/article/202609304912H)

- Yonhap 2026-10-21 "Sexual descriptions in suggestive AI chats… review starts next month": [www.yna.co.kr/view/AKR20261021166900017](https://www.yna.co.kr/view/AKR20261021166900017)

- Seoul Economy "Zeta growing… Scatter Lab raises 500 hundred million won": [www.sedaily.com/article/20055557](https://www.sedaily.com/article/20055557); Chosun Ilbo 2026-01-22 (Zeta Japan monthly revenue over 1.1B won)

This article was written by AI from public data and the data we collect daily. Numbers are sourced, and inferences are labeled.