The dataset started as riddles.

I am going to tell you the origin story of Eve-Genesis honestly, because the honest version is more useful to a clinical purchaser than the polished one. Eve-Genesis did not begin as a synthetic clinical reasoning corpus designed to train domain-specialised Small Reasoning Models. It began as a riddle dataset.

The intuition was simple. A riddle is not a knowledge test. A riddle is a cognitive test. It asks you to decompose, to trace implications, to resolve paradox. Riddles isolate reasoning from content. Train a Small Reasoning Model on riddles well and you have taught it reasoning style independent of any specific domain's knowledge.

The recognition

The breakthrough came when my daughter saw the dataset and named the formal logic underneath it. She threw words at me I had not been using: deductive, inductive, abductive. She had recognised, in a dataset I had designed under another name, the formal categories of reasoning that philosophy has been working with for centuries. Crucially, she named abductive reasoning — inference to the best explanation. That is the mode of differential diagnosis.

Confirmation came in a long conversation with an AI. The response named the layers in detail: dialectical, hermeneutic, semiotic, phenomenological, Socratic. And it noted, specifically, that the clinical chain I had been describing was abductive reasoning applied to medical signal. The dataset was structurally doing what classical clinical reasoning has been doing for as long as the discipline has existed.

The reframing

That recognition reframed the whole project. The dataset was not training the model on what diagnosis goes with what presentation. It was training the model on how to move between observation and explanation. That is a fundamentally different kind of training intervention. Most fine-tuning corpora carry input-output pairs. The model learns to associate inputs with outputs. Eve-Genesis carries reasoning structure as data. The model learns the operation itself.

The technical phrase is reasoning-style conditioning. The clinical phrase is the cognitive operation of differential diagnosis. Either way, the outcome is the same: the reasoner that emerges does not just know more medicine — it thinks like a clinician.

From riddles to clinical practice

Eve-Genesis (Clinical Edition) inherits the riddle-derived reasoning substrate and adds a layer calibrated to clinical cognition. The Clinical Edition emphasises two modes primarily: abductive reasoning (the substrate of differential diagnosis) and analogical reasoning (the substrate of case-based pattern recognition). Those modes are not generic AI capabilities. They are how clinicians actually reason.

Why the origin matters

I tell this story because the origin matters for the credibility of the claim. The riddle dataset was built before we knew what we were building. The dataset's structure was philosophically rigorous before anyone named it that. That is how a moat usually starts — not from a deliberate strategic decision to differentiate, but from following an intuition until it leads somewhere unexpected.