Responsible AI
Powerful when it helps. Restrained when it doesn’t.
We believe good technology is honest about its own limits. Here’s how Matchbook approaches AI — and why transparency is a feature, not an afterthought.
AI only when it adds real value
We don’t use AI for its own sake. It’s reserved for tasks where it genuinely improves your results — like understanding a garment from its images and description — not for decisions that simpler, transparent logic can handle better.
Rules-based logic where possible
Much of Matchbook runs on clear, inspectable rules that map your plain-language answers to styling attributes. This makes the system predictable, testable, and easy to explain — and keeps it fast and low-cost.
Analysis is cached, not repeated
When AI does analyze a product, we cache the result so the same item isn’t processed over and over. That avoids unnecessary AI calls, reduces energy and cost, and keeps the experience quick.
You stay in control
You choose what to share, you can shop any wardrobe you like, and “Not for Me” lets you steer recommendations directly. Your profile is yours to edit or clear at any time.
Recommendations are explainable
Every suggestion comes with a plain-language reason: why it works for you, how it fits, and its style vibe. If we can’t explain a recommendation, we don’t make it.
Our commitment
We design Matchbook so that you can always understand what it’s doing and why. We minimize the data we collect, avoid selling personal information, and prefer the simplest approach that produces a great recommendation. As we add capabilities — including future photo-based styling — we’ll hold them to the same standard: useful, transparent, and firmly under your control.
Questions about our approach?
We’re happy to talk through it.