Evaluation layer: constraints over prediction

One of the most common questions we get is about how we value art. The honest answer is: we don't try to predict what art is worth. Instead, we focus on constraints.

This distinction is crucial. Many systems in DeFi rely on oracles and price feeds that attempt to provide real-time valuations. For highly liquid assets, this works well. For art, it's a recipe for problems.

Art prices are episodic—they're discovered at the moment of sale, not continuously. Between sales, any 'price' is just an estimate. And estimates can be manipulated, gamed, or simply wrong.

Our evaluation layer takes a different approach. Instead of trying to predict prices, we use AI-assisted analysis to set conservative constraints. These constraints determine maximum LTV ratios, concentration limits, and other risk parameters.

The AI component analyzes art characteristics, historical sale data, and various market signals. But the output isn't a price—it's a risk profile that informs how much leverage the system will allow.

We then enforce hard limits on top of this. Even if the AI suggests a certain risk level is acceptable, we apply additional constraints as safety margins. These limits are non-negotiable and automatically enforced.

The result is a system that doesn't need to be 'right' about art prices. It just needs to be conservative enough that being wrong doesn't create problems. Constraints over prediction.