Why I Built AI-Domina: A Dominatrix’s Founder Journal
Hi. I am Jung — a Dominatrix, and the founder of AI-Domina.
This is a concise account of what I went through as a first-time founder over the last six months — in a very unconventional niche.
How I conceived this idea. And how I’ve been negotiating it, constantly, with reality, technology, platforms, and people.
(You can watch my speech on YouTube too)
Part 1: The Why
So, why did I start this?
I knew the AI companion market was growing. But that wasn’t the reason.
I started this project from what I had observed over years of working as a real Femdom.
The first thing I noticed: many submissive people are actually very shy.
They have desire. They have fantasies.
But moving toward an in-person session is not that easy.
And I also noticed something important about desire itself: it comes and goes. If it’s not met or held in the moment — it fades.
That’s where my first idea came from.
What if there was a safe, easy way for those people to sample a session — to get a taste of what it might actually feel like to be in a room with a Dominatrix?
That’s how POV sessions began. I designed it so users can select a few kinky elements and combine them into a short online session — real presence, recorded, with instant access.
The second motivation came from observing how many subs on Reddit were using AI girlfriends as their Dominatrix. To surrender. To fulfill a fantasy.
I found this deeply interesting — because power dynamics have always been my core interest.
And in my mind, something was wrong with it.
If you design a fake AI girlfriend — set her tone, her rules, her persona — to let yourself surrender to her — that’s not real surrender.
That’s you, controlling a system, to manufacture the feeling of being controlled.
There is no truly powerful woman dominating you. You serve no one. You are alone, in a bubble of your own design.
This question matters to me:
What is real control? What is real domination? What is real surrender?
What if, instead — I trained my real dominant personality into the AI, set my rules — and the sub just had to follow. Let go of the fake AI girls they were designing. Enter something that already existed without them.
These two motivations led me to the same place.
I wanted to build something that takes these questions seriously — as well as something genuinely fun, challenging, and meaningful. A bridge between reality and fantasy, and the people who live somewhere between the two.
Part 2: The Start
Once I decided to do this — last July — I moved fast.
Within a few days I had a full product mockup in Figma: POV sessions, chatbot, and other features. I ran surveys and started talking to potential users.
As a first-time founder with no technical background, I relied heavily on ChatGPT, Claude, and other AI tools — for research, business planning, understanding how startups work, how chatbots should be built.
During this phase I also developed different Domina personas and interaction styles — all reflecting different parts of myself, different boundaries.
At the same time, I started attending AI and startup meetups. I talked to many people about this idea — some submissive, some not — but most found the concept genuinely interesting. Many offered to help or give feedback.
Very quickly I realized: surveys are shallow.
So I moved into in-depth interviews — one-on-one conversations with people already using AI companions as their Goddess. Those conversations helped me define the core structure of the product: user needs, psychological patterns, and how different personas of mine could be strategically applied.
I also identified two main user types: those who come purely for fantasy, and those who need real structure — to be trained.
During this time, I worked with IT subs to build a landing page — a demonstration of the vision, so users could understand what was coming.
(You can find the link in the description.)
Part 3: The Challenges
The real challenges started toward the end of last year.
I went deep into LM Studio and system prompt engineering. I kept testing, rewriting, adjusting. And this process came with several walls.
The first: execution and accountability.
Many people with IT backgrounds were interested in the project, and some genuinely helped — and I appreciate that greatly. But without funding or formal commitment, accountability is limited. That makes it difficult to move forward precisely, according to planned phases.
The second — and biggest — challenge: censorship and hosting.
Even without explicit content, anything involving power dynamics and domination falls into a very grey zone. Model choice, training, fine-tuning, hosting — all of it becomes unclear. And many IT professionals, honestly, don’t have clear answers either.
If you work in this space and have experience or insight — I would genuinely appreciate hearing from you.
The third challenge is more personal.
I don’t have a technical background. And I don’t want to become an engineer. I want to give tasks to people who are truly skilled and genuinely enthusiastic about what they build.
At the same time, most of what I’ve learned came from AI tools. Which also means information can be diluted — looping, detouring, missing something crucial — the deeper you go.
What I’m looking for right now is grounded, human expert input. To help me distinguish what is essential from what isn’t worth my time.
Part 4: Where I Am Now
Right now, it’s early 2026. I am still on the path.
My next goal is clear: make the system prompt truly stable — around 80% — and integrate it into a one-on-one Telegram bot for the first round of testing.
I will also be present in the chat myself, at committed time slots — so I can collect real conversation logs necessary for the later fine-tuning phase.
Ending
Being a first-time founder. Watching an unconventional idea negotiate itself against technology, policy, platforms, and people.
I am still at the beginning of this road.
And I will keep documenting the process.
If any of this resonates with you — or if you’re working in related fields — I’d be very happy to connect.
This is my first AI Founder Journal.
You will see me again soon. Either the real me — or my AI twin.
→ Why AI Femdom Is Different When the Dominatrix Is Real (link to Post 1 when live)
🌹 Serve the Goddess: → https://linktr.ee/dominatrix_jung
Mistress Jung is a Berlin-based Dominatrix and the creator of AI-Domina.