I built a personality tool that is not a personality tool. I wrote that line months ago when I first put the Multi-Perspective Profile online. Four AI-guided conversations, each modelled on a real thinker, each designed to surface a different layer of who you are. Then it synthesizes what all four saw. It worked. People tried it. Something landed.

And then I made the mistake of asking AI what to do with it.

The question that started everything

The question was simple: could I commercialize this idea?

Not because I wanted to get rich. I wanted to know if what I had built was real — if it had enough substance to become something bigger. So I did what a lot of people do in 2026: I asked AI. Multiple AI tools. Multiple conversations. Multiple directions.

That is where things got interesting.

Five conversations, five different visions

One AI told me to build a “Social Navigation System” — a tool that maps ideological divides, bridges political polarization, and uses 3D visualization to let people fly through valleys of shared belief. It suggested passive profiling based on people’s browsing habits and a “Cognitive Diversity Score” with gamification badges.

Another told me to position the tool as a “translation layer between personality frameworks” and sell it B2C at 29 euros. Input a CV, answer a short questionnaire, get a unified personality narrative. Clean, practical, buildable.

A third said to add four sociopolitical lenses — Social, Political, Cultural, Ethical — and measure where people sit on belonging, authority, tradition, and fairness. It suggested a fifth “Epistemic Humility” lens to measure how certain people are that they are right.

A fourth said to hook the tool into people’s actual behavior — verify their stated Ikigai against their search history, compare their strategy style against their team’s average, and generate a “Life-Alignment Score.”

And the fifth — the one I ended up working with most — said: stop. You have enough inputs. You are not lost. You are just listening to too many voices that are optimizing for sounding exciting rather than being buildable.

That last one stung a little. But it was right.

What AI does well (and what it really does not)

Here is what I learned from running the same idea through multiple AI tools: each one will confidently build on whatever direction you lean toward. If you ask “could this be a social navigation system?” it will say yes and design you one. If you ask “should I add political lenses?” it will say yes and map them out. If you ask “what about passive profiling?” it will say yes and architect the surveillance infrastructure.

AI is extraordinarily good at saying yes and making the yes sound smart.

What it is not good at is saying “that is a completely different product than what you built, and you need to pick one.” It took a specific conversation — one where I deliberately asked for pushback — to hear that.

The feedback that actually changed my thinking was blunt. The sociopolitical lenses? “That is a worldview mapper, not a self-knowledge tool. Different product, different audience. The moment you measure someone’s political lens, you alienate half your users.” The passive profiling? “That is a surveillance product. It requires massive data infrastructure, raises GDPR issues, and is the opposite of what makes your tool interesting — which is that people actively participate in their own self-discovery.” The 3D visualization? “That is demo-ware, not a product. Nobody is going to fly through valleys of shared belief.”

I did not love hearing any of that. But it cleared the noise.

The rebrand that forced clarity

The first real decision was rebranding. The Multi-Perspective Profile had four guides: The Healer (inspired by Gabor Maté), The Provocateur (Esther Perel), The Clarifier (Simon Sinek), and The Philosopher (Yuval Noah Harari). They worked well for a blog post explaining the concept. But for anything beyond that, they created problems.

You cannot trademark someone else’s name. You cannot build a product around borrowed authority. And the moment you charge for a tool that says “talk to Esther Perel,” you are one cease-and-desist away from a very bad week.

So the names had to go. What replaced them was, honestly, better.

The Root — what is underneath your choices. The Mirror — what you are not seeing about yourself. The Thread — what keeps coming back across your career. The Orbit — the story above your story. Four lenses, each with a clear function, each standing on its own without needing a famous name attached.

The name “Convergence Profile” came from the core idea: the value is not in any single conversation. It is in the convergence — where multiple perspectives overlap and the real signal emerges. When three lenses see the same pattern, that signal is strong. When they contradict each other, that tension tells you something too.

The gap I did not want to find

Then came the uncomfortable question: are four lenses actually enough?

I asked for a proper validation — not “does this feel right?” but “map these four lenses against what established psychology says a complete profile needs to cover.” The answer came back with twelve essential dimensions from the research literature. My four lenses covered roughly 40 to 50 percent.

The biggest gap was behavioral. The Root, Mirror, Thread, and Orbit are all strong on the introspective side — emotional drivers, identity, purpose, narrative. But none of them directly ask: how do you actually behave with other people? What happens to you under pressure? What role do you default to in a group? How do others experience you?

That is when The Field was born — the fifth lens. “How you show up.” Observational, scenario-based, focused on what you do rather than what you feel. It closed the gap between inner experience and outer reality, and brought coverage up to about 75 to 80 percent. The remaining 20 percent — cognitive processing style and clinical mental health markers — is genuinely outside the scope of a conversational tool. That is an acceptable limitation, as long as you are honest about it.

The assessment stack problem

The original tool mapped output to Big Five, DISC, CliftonStrengths, and Ikigai. I chose them because I knew them. That is not a good enough reason.

So I asked: are these actually the best frameworks, or just the ones I am familiar with?

The research turned up one critical finding that none of the other AI conversations had flagged: CliftonStrengths is trademarked by Gallup. All 34 theme names are proprietary. Gallup actively enforces their IP. You cannot map output to CliftonStrengths in any product — even a free one — without written permission.

That alone forced a change. CliftonStrengths was replaced by VIA Character Strengths — 1,000 peer-reviewed studies, developed by 55 scientists, and the VIA Institute actually encourages use. Same purpose, better science, no legal risk.

The research also surfaced a missing dimension: values. The Schwartz Values framework has been validated across 82 countries and captures something none of the other frameworks do — what fundamentally matters to you. That is different from personality traits, different from strengths, different from purpose. It is your priority system. It was added.

DISC stayed, but with a demotion. The science behind it is weak — it is basically a simplified version of Big Five that loses nuance. But people understand it instantly and apply it at work. It earns its place as a practical communication layer, not as science.

And the Tension Meter — the element I had built on instinct, placing it at the bottom of the profile as an afterthought — turned out to be the most distinctive part of the entire output. Every AI I consulted flagged it as the differentiator. No standard assessment tool maps where your strengths pull in different directions. Clarity versus Complexity. Autonomy versus Structure. Depth versus Speed. The tensions are where the real insight lives.

It moved from the bottom of the profile to the top.

The moment I almost lost the plot

There was a point in this process where I had a personal assessment tool, a sociopolitical worldview mapper, a social navigation system with 3D visualization, a passive profiling engine, and a burnout prediction dashboard. All supposedly the same product.

The feedback that snapped me out of it was this: “You have been exploring this concept with multiple AI tools, and each one has taken it in a different direction based on what sounded exciting in that conversation. They are all interesting individually. But they cannot all be the same product. You need to pick one.”

And then: “The next insight you need will not come from another AI analysis. It will come from someone actually going through the conversations and telling you what landed and what did not.”

I think that is the most important thing I learned through this entire process. AI is spectacularly good at generating options and making each one sound like the right one. It is not good at telling you when to stop generating and start building. That part is still yours.

What I ended up with

The Convergence Profile. Five lenses: Root, Mirror, Thread, Orbit, Field. Each surfaces a different layer. Users choose a guide they connect with — the methodology runs underneath. The output synthesizes across Big Five, VIA Character Strengths, Schwartz Values, DISC, and Ikigai. The Tension Meter sits at the center. It is free for individuals.

One sentence: five conversations that show you what no single personality test can.

Is it the best possible version? No. It never will be. But it is a locked-down version one that I can put in front of real people and learn from their reactions. And that is worth more than another round of AI-generated possibilities.

What I would tell someone starting a similar process

Use multiple AI tools. Seriously. The diversity of perspectives is valuable. But know that each tool will enthusiastically build on whatever direction you point it in. That is a feature and a bug.

Ask for pushback explicitly. Most AI defaults to agreement. If you want honest feedback, you have to request it — and even then, you have to push back on the pushback to see if it holds.

Watch for the moment when you have more visions than you can reconcile. That is not a sign that you need more input. It is a sign that you need to decide.

And know that the most useful AI conversation is not the one that gives you the most exciting idea. It is the one that helps you throw away the four ideas that were distracting you from the one that actually works.

The Convergence Profile exists because I eventually stopped asking “what could this be?” and started asking “what is this, right now, and is it good enough to share?”

It is. So I am sharing it.