Apple has recently acquired Silk Labs, a startup which focused on building on-device machine learning software. The purchase, which took place earlier this year according to The Information, would be a natural fit for Apple’s privacy centric approach to AI. After an unsuccessful foray into the smart home, Silk Labs developed on-device machine learning services. This means processing data without sending it to the cloud a method that Apple has also preferred.
Silk Labs itself is not a familiar name in tech. The company was co-founded by former Mozilla CTO Andreas Gal and launched only one product an intelligent camera and hub for your smart home name Sense. The hub was stylish and well-designed, and intended to connect together smart home products including thermostats, speakers, and cameras.
Sense was launched on Kickstarter in February 2016 but received only around $150,000 in funding from 774 backers. Silk Labs cancelled the product in June that year refunding backers, and announced that it was pivoting to developing AI software for other firms.
The company’s website shows a range of services that draw on machine vision technology. These include people detection and facial recognition for tasks like building surveillance, retail analytics, and home security. On the company’s site and on social media, it stresses that its use of edge computing and encryption make its services more private than those of rivals.
Apple has taken a similar approach in its development of AI, differentiating itself from rivals like Google, which collects substantial amounts of user data and processes it in the cloud. When it comes to performing analysis of your data, Apple exec Craig Federighi said in 2016: “We’re doing it on your devices, keeping your personal data under your control.”
According to The Information, the purchase price for Silk Labs was likely small. The startup only had around a dozen employees and had raised approximately $4 million in funding. We’ve reached out to Apple for comment and will update this story if we hear more.
