Apple is using machine learning technology in more and more of its features, John Giannandrea, the company’s senior vice president for machine learning and AI strategy, reveals in a new interview with Ars Technica.
Giannandrea, who joined Apple from a job at Google, said “there are increasingly fewer and fewer places in iOS where we’re not using machine learning.” This stands in strong contrast to a few years ago — and even the point at which Giannandrea landed his Apple job.
Apple and machine learning
“When I joined Apple, I was already an iPad user, and I loved the Pencil,” Giannandrea said. “So, I would track down the software teams and I would say, ‘Okay, where’s the machine learning team that’s working on handwriting?’ And I couldn’t find it.”
According to him, the team didn’t exist — despite the fact that machine learning was vital for state-of-the-art handwriting recognition. Jump forward a couple years, and handwriting recognition is now a part of the upcoming iPadOS 14.
“I knew that there was so much machine learning that Apple should do that it was surprising that not everything was actually being done,” Giannandrea said. “And that has changed dramatically in the last two to three years. I really honestly think there’s not a corner of iOS or Apple experiences that will not be transformed by machine learning over the coming few years.”
Giannandrea said machine learning is already being used by Apple in a number of areas. These include language translation, or on-device dictation, and new health-related features like sleep-tracking. He continued that Apple’s augmented reality efforts are also reliant on machine learning for things like lidar-based mapping. Overall, he suggests that Apple’s fairly unique position of designing both the hardware and software for its devices means it is well-positioned to “lead the industry” in this domain.
Catching up on AI
For a long time, Apple lagged its rivals on artificial intelligence research. Things changed in 2015 when Apple began to ramp up hiring in this area. Machine learning has played an increasingly large part in Apple products since then. In 2017, Apple launched its own quasi-academic machine learning journal, allowing Apple AI researchers to publish their work.
Crucially, Apple has focused on finding ways to do smart machine learning without having to sacrifice users’ data privacy. On-device AI isn’t unique to Apple, though. Other companies, including Google, work on this type of thing.
Are you impressed by Apple’s artificial intelligence efforts? Let us know your thoughts in the comments below.
Source: Ars Technica