Apple’s New AI Era: Sensational, But Not for Every iPhone User
In the tech world, a watershed moment is nigh with Apple's forthcoming artificial intelligence (AI) system, aptly dubbed Apple Intelligence. Promising an entire makeover of the user's interaction with iPad, Mac and iPhone, this AI innovation is set for launch before the year is out.
Revolutionary or Evolutionary?
Apple Intelligence, critics say, might not break entirely fresh grounds but is likely to end up as one of the most user-friendly, intuitive and practical applications of generative AI seen so far. Enthusiasts are excited about its transformative potential with applications ranging from refining Siri's smartness, crafting unique emojis, improving emails and essays, to turning rudimentary sketches into computer-generated artwork.
Restricted Access
Notwithstanding the buzz, there's a catch - the state-of-the-art AI will only be available to users of iPhone 15 Pro or iPhone 15 Pro Max. Older versions, including the scarcely a year-old regular iPhone 15, miss out on these features. Mac users will also need a device released in 2020 or after to tap into its capabilities.
Challenging Computations
The disruptive nature of Apple Intelligence comes with underpinning complexities. The required computing operations for Apple Intelligence differ significantly from typical tasks performed on iPhones or Macs. Thus far, users have been largely shielded from these complexities since generative AI processes have been largely cloud-based, conducted on remote servers.
Different Take on AI Implementation
In a shift from the traditional model, Apple Intelligence aims to execute some part of these operations on the gadget itself, decreasing reliance on external servers. Craig Federighi, Apple's senior VP of Software Engineering, elaborates on this during the AI announcement, "We have integrated it deep into your iPhone, iPad, and Mac, and throughout your apps, so it’s aware of your personal data without collecting your personal data."
AI Processing and Privacy
The on-device AI processing, apart from offering benefits in terms of performance, is also a strategic move for privacy protection - a staple aspect of Apple's philosophy. However, Apple isn't alone in this field. Microsoft's Copilot+, a similar concept focusing primarily on laptops, also utilizes locally based AI processing.
Hardware Requirements for AI
At the heart of the issue lies the innovation of the "neural processing unit" (NPU), pivotal for AI's functionality. The NPU is an add-on to the typical CPU and GPU found in devices, focusing on performing many operations concurrently with minimal power consumption, crucial for portable devices like phones and laptops.
NPU and AI History
For context, Apple introduced the NPU as far back as 2017 with the release of the iPhone 8 and X models. Forms of AI and machine learning have been a part of iOS systems for several years, but what differentiates Apple Intelligence is its concentrated focus and possible artificial gating of these "AI" features to newer hardware.
The Role of Memory
The factor that distinguishes gadget variants that are compatible with Apple Intelligence might not be AI capabilities per se but the amount of Random Access Memory (RAM) available. For on-device AI processing, models need to be stored in RAM as disk storage speeds don’t suffice. Available models of RAM-compatible iPhones start from 8 GB, leaving older ones with just 6 GB out of the party.
The Future of AI-enabled iPhones
The bottom line remains that while the AI innovation in itself seems groundbreaking on paper, its practical implementation on most hardware could be limited due to spatial and computational constraints. As anticipation builds for iOS 18, set to unveil Apple Intelligence, users will eagerly watch whether these lofty AI features will become a staple of even the non-pro "iPhone 16" or continue to remain confined to the more exclusive Pro series.
