Progressive Disclosure

Progressive disclosure is the single most important principle in information architecture. Show the minimum needed to make a decision. Reveal more when the user asks for it. Never dump everything at once. It sounds obvious. Almost no AI product does it.

Layers of Detail

A Persephonie tree has natural layers. The root question is layer zero. Primary options are layer one. Outcomes are layer two. Deep analysis is layer three and beyond. Each layer is a complete, useful view. You never need to go deeper, but you always can.

Compare this to a chat response that gives you everything at once. The key insight is buried in paragraph four alongside caveats, edge cases, and tangential context. There are no layers. There's no prioritization. The important and the trivial sit side by side.

The first layer should answer the question. Every layer after that should answer 'tell me more.'

On-Demand Complexity

• Layer 0: the question, always visible as the root
• Layer 1: primary options, visible immediately
• Layer 2: outcomes and trade-offs, one click away
• Layer 3+: deep analysis, available for those who want it

Trust Through Transparency

Progressive disclosure builds trust. When users see a clean, simple first layer, they trust the interface. When they click deeper and find thoughtful detail, they trust it more. The opposite, dumping everything at once, signals that the interface doesn't know what's important. And if the interface doesn't know, why should the user trust it to?