Apple is doubling down on artificial intelligence that stays on your device, not in the cloud. At its September hardware event, the company showcased how its new Siri AI and on-device machine learning features will work across the iPhone 18 Pro, iPhone Duo, and Apple Watch Series 12, marking a significant shift in how the tech giant approaches AI compared to competitors who rely heavily on cloud processing (Source 1, 2, 3).

What Makes Apple's On-Device AI Different?

Apple's approach to artificial intelligence centers on a fundamental principle: keeping your data on your device. Rather than sending conversations, health data, or personal information to remote servers for processing, Apple's new Siri AI and machine learning features run directly on the phone or watch itself. This strategy addresses growing privacy concerns while delivering faster responses since data doesn't need to travel to distant data centers (Source 1, 2).

The company emphasized this privacy-first approach during its keynote, positioning it as a key differentiator from rivals who are further along in incorporating cloud-based AI. Apple's new Neural Engine on the A20 Pro chip has 32 cores, enabling sophisticated AI processing without sacrificing privacy.

How Does Apple's New Siri AI Work on Your Devices?

The updated Siri AI brings personal context understanding and broad world knowledge to Apple's ecosystem. On the Apple Watch Series 12, Siri will power three new audio-focused features that demonstrate the power of on-device processing. These capabilities show how machine learning can enhance everyday interactions without requiring cloud connectivity.

• Live Rewind: By double-pressing the Digital Crown, users will be able to see the previous 15 seconds of a conversation displayed as text, allowing them to catch something they missed or didn't fully understand without needing to ask the other person to repeat themselves.
• Siri Recap: After a conversation, meeting, or lecture, Apple Intelligence will automatically generate a title and key points that users can review later, helping them stay present in the moment while still capturing important information.
• Sound Recognition: The watch listens for important sounds like sirens, alarms, doorbells, or a baby crying, using on-device intelligence to alert users who are deaf or hard of hearing even when their iPhone isn't nearby.

Live Rewind and Siri Recap run on Siri AI. Sound Recognition uses on-device intelligence. All are designed with privacy and security in mind. Apple stated that audio is never recorded, nor can the system access the audio itself, and speakers are not identified. All data that syncs is encrypted by default (Source 1, 2).

Why Is On-Device AI Processing Important for Users?

On-device machine learning offers several practical advantages beyond privacy. First, it eliminates latency caused by sending data to remote servers and waiting for responses. Second, it works even when you don't have an internet connection. Third, it reduces the risk of data breaches since sensitive information never leaves your device. Apple's new A20 Pro chip, built on 2-nanometer technology, includes a seven-core GPU with 40% faster graphics performance and a Neural Engine specifically optimized for these machine learning tasks.

The company's Health Sensing System on the Apple Watch Series 12 exemplifies this approach. The watch takes continuous heart rate readings every five seconds, which is 60 times more frequently than previous Apple Watches. It also checks heart rate variability every five minutes, 24 times more frequently than other watches on the market. All of this processing happens on the device itself, meaning your health data stays private while you get real-time insights (Source 1, 2).

How to Use Apple Intelligence Features on Your Devices

• Control Audio Intelligence Settings: Users can enable or disable Sound Recognition through their watch settings, giving them full control over which on-device AI features they want active. This feature is available now on Apple Watch Series 12.
• Prepare for Live Rewind and Siri Recap: These features will ship in beta later in 2026. When available, users will be able to double-press the Digital Crown during conversations to see text transcripts, and review automatically generated summaries of meetings and lectures in the new Siri app.
• Monitor Health Insights with Apple Intelligence: The redesigned Health app now includes Apple Intelligence summaries and recommendations based on your continuous heart rate data, activity levels, and sleep patterns, providing personalized wellness guidance available today.
• Check Your Health Age and Longevity Metrics: The new Health app features a "Health Age" tool that gives you advice based on what your data says about your biological age, and a Longevity tab that assesses your overall health across multiple metrics.

The iPhone 18 Pro and iPhone 18 Pro Max also benefit from this on-device AI approach. The A20 Pro chip delivers a 50% increase in memory bandwidth compared to previous generations, enabling faster machine learning processing. The phones support up to 36 hours of video playback for the Pro model and 45 hours for the Pro Max, with wired charging reaching 50% battery in just 15 minutes.

Apple's new S11 chip in the Apple Watch Series 12 processes all these Apple Intelligence features with a dual-core CPU, single-core GPU, and quad-core Neural Engine derived from the A20 Pro. The Series 12 starts at $399 and offers up to 24 hours of battery life, while the Ultra 4 provides over two days of battery life and starts at $799.

What Do Experts Say About Apple's AI Strategy?

"The foldable launch is straight out of Apple's playbook: Wait for the product to work out its kinks and for the market to offer glimpses of viability, then enter and shape the category," noted Dipanjan Chatterjee, analyst at Forrester.

Dipanjan Chatterjee, Analyst at Forrester

Chatterjee's observation about Apple's approach to new product categories reflects the company's broader philosophy of entering markets thoughtfully rather than rushing. This same deliberate strategy extends to how Apple positions its AI capabilities, prioritizing on-device processing and privacy over being first to market with cloud-based AI solutions.

The timing of these announcements coincides with John Ternus taking over as Apple's new CEO on September 1, 2026, succeeding Tim Cook. Ternus used the keynote to emphasize Apple's commitment to privacy and security as differentiators in the AI space, signaling that on-device machine learning will remain central to the company's vision going forward.

Apple's focus on on-device AI represents a deliberate choice to compete on privacy and user control rather than raw processing power or feature speed. As the company rolls out iOS 27 with these new Siri AI capabilities, users will see how machine learning can enhance daily life while keeping personal data secure and local (Source 1, 2).