Workflow 3
Live in beta

Overseas Adaptation: new skin, same bones — Chinese hits become local-market shows

Three confirmations — genre, market, reference show. Then skeleton inheritance plus reskin assembly: emotional anchors and pay-point structure are replicated episode by episode, while characters, worldbuilding, and cultural hooks are rebuilt for the US, Brazil, or Indonesia.

For: short-drama studios going global, international content labels, producers scaling ROI on an existing catalog

How it works

Done in 4 steps

  1. 1

    Confirm the genre

    Upload a finished Chinese script; AI identifies the genre, maps it to genres that sell overseas, and grades adaptability.

  2. 2

    Pick market & reference

    Choose the US, Brazil, or Indonesia; AI recommends worldview directions and benchmark shows from per-market cognition matrices.

  3. 3

    Inherit the skeleton, assemble the skin

    Emotional anchors and pay points are inherited episode by episode; the worldview package is assembled and frozen on your sign-off.

  4. 4

    Adapt episode by episode

    Localized episodes are produced under the target platform's paywall rules, with bilingual dialogue and full traceability.

Core capabilities

Why it's different

New skin, same bones

The hard asset in going global is pay-point repositioning, not translation. The macro outline (characters / acts / setups) stays frozen, per-episode emotional anchors are replicated, and concrete plots and settings are rebuilt from the ground up.

Evidence from 30 shows and 1,419 episodes

The structural rules aren't imagined: they come from episode-level teardowns of real overseas hits (25,535 action lines / 28,643 dialogue lines) — first paywall, act structure, hooks, all sourced.

World assembly with a naming hard gate

Factions, landmarks, names, and full character cards are assembled from worldview template libraries, with genre-grammar consistency checks and a hard gate on character naming — no more "Chinese names in Western suits".

A copyright clean room

Three layers: legal source material → abstracted patterns → AI-original generation. The product never outputs copyrighted text; spot checks and keyword scans return zero hits.

Target markets

Three main battlegrounds, one cognition matrix per market

Editorial knowledge like "US audiences won't take a tragic romance without a hard reversal" is structured into a per-market genre-fit matrix — every adaptation direction clears the market gate first, and editors' rejections keep flowing back as calibration.

United States

English

The largest overseas short-drama market. Werewolf, billionaire, and hidden-identity genres carry paywall structures markedly different from China's — repositioning pay points is the real work.

Brazil

Portuguese

One of the fastest-growing markets, with a taste for high emotional intensity. Localization goes past translation into forms of address, religion, and family structure.

Indonesia

Bahasa Indonesia

The core Southeast Asian market, highly sensitive to cultural taboos. The market profile carries built-in taboo filters and steers around religiously charged elements automatically.

Measured in production: producing an overseas variant of a finished Chinese show went from one translator plus one local editor over 2–3 weeks down to 3–5 days. A light rename pass costs roughly 1/20 of a full reskin, so one source script yields multi-market variants cheaply.
FAQ

About Overseas Adaptation

How is this different from machine translation plus a local polish pass?
Translation moves the language; we move the structure. Overseas platforms place paywalls and first walls differently from China — the pipeline recomputes pay-point coordinates first, rebuilds the worldview and cultural hooks for the target market, and translation is only the final layer. In production this compressed 2–3 weeks of variant production into 3–5 days.
What about copyright risk?
A three-layer copyright clean room: only legally public sources are collected, everything is abstracted into patterns at intake, and the output layer is entirely AI-original — copyrighted source text is never emitted; reference shows inform structure only.
Which languages and markets are supported?
Currently the three main battlegrounds of Chinese short-drama exports: the US (English), Brazil (Portuguese), and Indonesia (Bahasa Indonesia) — each with its own market profile and genre-fit matrix, expanding market by market.

Run your next script through the data first

The invite-only beta is open. Tell us about your team and use case — we'll get you set up shortly.

Every shipped capability is free to use during the beta