We don't want to build one more "type a line, get a line" tool, where a human still has to bring the real insight and the machine only polishes the surface. We're building a B2B custom narrative engine that starts earlier: it reads what a crowd is carrying, then turns that pressure into story maps, motif sets, world bibles and scene boards.
To put it precisely: we're not designing a tool. We're extracting stories from audiences, before production begins. This is Project Thalassa.
Most AIGC tools are still downstream tools. They respond to prompts, rewrite scripts, render images, score drafts or predict whether something might work. Useful, but they don't solve the hardest question in early development: what story should exist in the first place, and why would this audience need it?
That gap is why we use the name Thalassa. It means the sea, the deep layer under the visible surface. The surface is content. Under it are anxiety, aspiration, social memory, class pressure, longing and shame. Thalassa is the layer where those submerged audience signals become narrative material.
The engine has three moves. First, observe: read audience signals at the group level, not individual profiles. Second, bind: connect emotions with recurring symbols, scenes and social tensions. Third, output: turn that into a narrative package a production team can use.
The point is simple. We are not asking AI to invent from nowhere. We are asking it to listen to the audience more deeply than a survey or a sentiment tag can, then return something structured: a story map, a motif set, a world bible, scene boards. That is the emotional engine behind the storytelling.
The first business form is not a consumer app. It is project-based B2B custom service. For a film, a series, a game world, a cultural tourism project or a brand campaign, the client comes with a brief. We add the missing layer: what story is already latent in the audience.
Take one word, "a flat". To someone who's moved to London, what is it all at once? The right to stay, the school catchment for their kids, the chance to stand taller when they go home at Christmas. And also a twenty-five-year mortgage that owns them, a ladder they can't climb back down. Those hopes and fears, tangled together, are the story pressure nobody says out loud. A sprint turns that pressure into a production-ready narrative package.
Technically, the one thing that matters is this: we compute motifs, we don't just tag feelings. The visual on the screen is the evolution end, where audience signals, symbols and emotional tensions meet, bond, drift and recombine like cultures in a petri dish. There are three ends in all: collection, evolution and output, with no individual profiling.
In simple terms, VSA, Vector Symbolic Architecture, lets symbols become high-dimensional vectors, so they can bind, separate and recombine mathematically. VSM, Vector Space Model, places emotions, motifs and story elements in one shared space, so we can read proximity, tension and drift.
The result is not an abstract dashboard. It lands as story maps, role tensions, world logic, scene boards and design directions, ready for a production team to discuss, test and iterate.
Foreign platforms give us a clear benchmark. Many are strong in testing, prediction, analytics or rendering. They help a studio judge the script, the audience fit, the market risk or the visual output. They are useful, but most of them start after the story already exists.
Thalassa sits one step upstream. We work before the story is fixed: extracting the emotional motif, the character pressure, the world premise and the scenes worth testing in the first place. Others help you decide. We help you originate.
This doesn't stay inside a slide deck. The core is turning audience emotion into narrative assets a client can actually use: a premise, a character structure, a world logic, a scene direction. The early commercial path is project-based custom service for cultural tourism, film and television, fashion, brands and export-facing stories.
What's on the screen is a demo of that direction. What it's testing is fine control of feeling: whether we can read emotional symbols from an audience and carry them back as a story that still feels alive to that audience.
First example, an AI series. We start with a chronicle, hundreds of millions of words, a parent world: settings, history, events, places, rules, all laid down so the world stands on its own first.
Then, from social symbols and a crowd's desire, we pull each character's motive, conflict and tension. A thick enough world gives a character a past; a real enough struggle inside them, and the story starts to come on its own. What's on the screen is the emergent narrative we're testing now: not AI writing one episode and stopping, but one parent world that keeps growing new stories out of itself.
Second example, a feature animation for cinema. We already have a script, and we're expanding its world into a whole universe. We want every character in it to be no function, but the kind with a past, with a struggle, the kind you remember for a long time.
Here's how: we use the world a virtual character dreams up to design the actual content, pulling the visuals, the spaces, the props and the atmosphere out of that dream into a design method. With it, even pre-production design can be produced at volume, every day.
The hard part of Thalassa is that it sits across storytelling, AI, psychology and semiotics. It won't stand if any one is missing. I'm Dai Shang, a PhD from Tsinghua University; Chunling Wu is a PhD from the University of London. The team structure is intentionally compact: research, narrative method, algorithmic prototyping and production handoff all have to stay close in the early stage.
What's next is three things. One, more audience-story data: crowd emotion, cultural motifs, worldview samples and commercial-scene cases. Two, tuning the algorithms so motif bonding and narrative generation become steadier, sharper and easier to explain. Three, B2B pilots in tourism, film, brands and games, letting each custom project flow back into a reusable narrative library.
Finally, the ask. What we need most now is not noisy scale; it is pilots, a tighter delivery team and a deeper narrative library. The money goes to audience-story cases, motif vectorisation, custom sprint delivery, and the engineering needed to keep story extraction and narrative generation sharp. We're open to a seed round, at a pre-money valuation of RMB 20 million.
Thalassa is named for the sea because every public story has a surface and a depth. The surface is what people say they like. The depth is what they need, fear, envy, remember and cannot quite name. Our work is to bring that depth back as story. Thank you.