Questions & Answers

Generative AI refers to a family of technologies that can create images, videos, music, or text based on written input — called a prompt. These tools don’t “imagine” anything on their own. They’re powered by complex models trained on billions of examples, and the outcome depends entirely on how precise, creative, and intentional the input is.

In my work, I use generative AI to build visual assets, scenes, transitions, environments, objects, even characters. But it’s far from a “click-and-done” process.

Each result involves multiple iterations, refinement, style balancing, consistency adjustments, and post-production work. The tools are still evolving, and current limitations include:

• Low frame rate or motion instability

• Difficulty with fine detail over time

• Limited control over physics or perspective

• Inconsistencies between scenes

• The growing number of platforms and tools — each with different strengths, weaknesses, and quirks

• The big difference between generating a simple image and crafting a consistent, refined visual style across a full project

• And yes — the real cost of generation time

That’s why I don’t just generate, I design. I build every piece like a puzzle: AI is just one part of a larger creative process that includes prompt mastery, aesthetic research, editing, sound design, narrative structuring, and visual direction.

Generative AI opens powerful new doors for creativity and yes, it’s faster than hand-drawing every frame. That’s part of what makes it more affordable. But if you think what you need can be done in a few minutes by anyone… feel free to give it a try, and let me know how it goes, I would be happy to ear your feedback . 😉

Hi, I’m Jonathan Carène, a French filmmaker and visual storyteller based in France I am fluent in English, Japanese, and Italian.

My journey began with a 1h39 documentary, “DEKIRU – Eco Solutions in Japan,” which I co-directed with my wife. The project was supported by the French Embassy and is now freely available online. This experience ignited a lasting passion for filmmaking.

Before launching Masayume, I gained hands-on experience in various fields: retail, food service, and event management. In 2010, I founded AOJI – The Art of Japanese Illustration, an online gallery dedicated to Japanese art, which I ran for five years. These formative years helped me develop solid, versatile working skills and a strong entrepreneurial mindset.

I later turned to filmmaking, gradually learning the craft of video production while working in the tourism and creative industries, always exploring, refining, and expanding my artistic vision.

Today, I’m a full-spectrum filmmaker. I imagine, script, shoot, and edit each project myself. I work with a Sony A7 III, premium lenses, and have been using DaVinci Resolve as my primary editing tool for over six years.

When I discovered the power of generative AI, I immediately saw its potential — not as a shortcut, but as a creative ally. After months of training and exploration, I developed a complete workflow that lets me combine AI-generated images, videos, music, and even custom voice work to build truly immersive stories. I also use Photoshop to fine-tune visuals and maintain full artistic control over each piece.

Every project I take on is unique. It carries a story, a vision, an emotion — and it’s always a pleasure to unfold them with my clients.

With Masayume, I merge real-world shooting, generative AI, design, and poetic storytelling to craft visuals that move people — emotionally and visually. I work solo most of the time, from concept to final delivery, with occasional collaborations when needed.

Everything you see here was built with curiosity, care, and a deep love for visual culture and digital creation.

Absolutely, I’m in love with art, and especially with Japanese animation, Franco-Belgian comic books, and of course, video games. I’ve always been drawn to powerful visual universes, and if I could, I’d work with real artists every single time.

But the reality is: at this stage, I simply couldn’t afford it. The cost of hiring illustrators, animators, musicians, and editors for every project would be too high and Masayume wouldn’t exist.

That’s why I use AI as a creative tool, not as a substitute for real talent. It allows me to explore, craft, and deliver custom visual stories with a unique touch even on limited budgets.

I’m sure that someday, I’ll be able to wield enough budget to work with talented artists and create hybrid masterpieces for my customers. That’s the vision I’m aiming for.

I know many artists feel uneasy about the current shift in visual content creation and I understand them completely. To be honest, I never expected this change to happen so fast. A part of me feels genuinely sad, knowing that the hard work and talent of so many artists might get swept away in a tidal wave of AI-generated content.

I believe artists are essential to our civilisation, and we need to protect them. That’s why I fully support international regulations that would forbid the use of an artist’s or studio’s style without their consent or compensation. Generative AI tools are trained on the work of millions of creators. The technology is here now — there's no stopping the wave — but using an artist’s name in a prompt is, in my eyes, theft. It must not be normalised.

In my work, I never use the name of an artist or studio when generating content, unless the project is directly related and the reference is ethically justified. I will always be transparent on that matter.

I may not draw, play an instrument, or carve wood — but I use these new tools with a deeply artistic approach. Every project I deliver is the result of a creative process that blends intention, style, and human sensitivity.

Generative AI is powerful — but it’s still a tool. While anyone can generate a pretty image, creating something coherent, original, meaningful, and well-crafted takes skill, experience, and a strong creative direction.