Standard Blog

David Valdman
David Valdman
Co-founder - Machine Learning
Earning Trust

Earning Trust

Recently, Oakland, CA became the third US city (along with San Francisco, CA and Somerville, MA) to ban the use of facial recognition technology. The city passed an ordinance prohibiting the use of facial recognition because the technology was inaccurate, potentially invasive and lacks standards. While this regulation doesn’t directly affect us as a company, we think it’s a great opportunity to discuss how this issue relates to our values and the decisions we’ve made in building our products.

First, we welcome this decision to ban facial recognition and believe it’s one step in an inevitable (and necessary) series of steps to regulate the amazing capabilities coming from deep learning. With technology becoming ever more powerful, a company needs to carefully consider its capabilities against civic interest, whether it be facial recognition, misinformation spread on social media, or something else. We see this recent regulation, along with the well-established GDPR law in Europe and the coming CCPA regulation in California, and say “Yes. And?”

In order for Standard to be successful, we need to earn the trust of both retailers and shoppers. Let’s be honest, our business is based on installing cameras into retail stores; we start with a trust deficit and a skepticism surplus. From that point forward, we know we have a lot to prove. Trust is earned not only by your actions, but by what you don’t do. We have a firm stance to not collect any biometric data from the shopper, of which facial recognition is a type. We take this stance knowing it’s technically more challenging. But if we need to choose between a technical shortcut and earning shopper trust, trust wins. Similarly, for retailers, we know using shelf sensors would make our technology simpler, but it puts a burden on installation and flexibility. Only by putting your users first can you build a great product.

It’s the little things that build trust too. We have the only autonomous checkout store, demo or otherwise, that doesn’t require an app to enter or a gate to cross. Shoppers without our app can walk up to our kiosk, which magically knows what they’ve purchased, and pay completely anonymously.

What’s true for our store will be true for the retailers we work with. We want to enable an effortless shopping experience. To do that, we actually don’t need to know much about shoppers. We also have a business model which isn’t ad-supported, so we don’t have the pressure other companies do to track their customers. In short, we’ve designed our business to not need to be creepy. This means technical tradeoffs and an indoctrination of our values into our products, from code and design to features and experiences.

Facial recognition may be the issue du jour, but to us it’s a stepping stone to a larger goal: earning trust. The ways to earn trust are to protect privacy, create transparency, invest in security and best practices, state what you do and don’t do, collect only the data you need, and delete the data you don’t. In the future there may be regulations for all of these things. Then, as now, we say “Yes. And?”

All blog posts

©2019 Standard Cognition
965 Mission St, Floor 7
San Francisco, CA 94103

Blog · Careers · Privacy Policy · Terms of Use