Apple Intelligence is a result of Apple’s comprehensive silicon initiatives, as revealed in a recent interview. Notably, these initiatives made a significant pivot back in 2017 to lay the groundwork for AI advancements.
The latest episode of The Circuit podcast showcases a dialogue between Ben Bajarin, Jay Goldberg, and Apple executives Tom Boger and Tim Millet.
Their discussion centers around Apple silicon, highlighting the intersection between AI and the company’s silicon developments.
We discover that the original Neural Engine was designed to enhance Apple’s computational photography capabilities, which ultimately positioned them to thrive in the AI space.
Here’s a notable quote regarding a 2017 decision that facilitated AI integration in M1 devices:
“We introduced [the Neural Engine] in 2017, but another significant event in 2017 was the publication of the paper, Attention is All [You Need]. This paper paved the way for transformer networks… My team was aware of this; they were reading the paper in 2017 and thought it looked promising. We knew we had to ensure our capability in this area.”
“Thus, we began re-architecting our neural engine the moment we started shipping it, so by 2020, when we launched the M1 during the Apple silicon transition, we could run these networks. This foresight allowed us to commit to implementing Apple Intelligence across all Macs powered by Apple Silicon, knowing that silicon development requires time.”
While Apple Intelligence was not on the radar in 2017, the proactive modifications to the Neural Engine by Apple’s silicon teams were essential to enabling 2020 Macs to support the latest software today.
The entire interview can be accessed on Apple Podcasts and is definitely worth a listen.
What are your thoughts on Apple’s insights regarding AI from the interview? Share your views in the comments.
Top Apple Product Accessories
FTC: We utilize income-generating auto affiliate links. More.
