Apple to spotlight on-device AI capabilities and custom silicon advantage at WWDC 2026: report

With WWDC right around the corner, new leaks and revelations keep on adding to the fevered pitch surrounding the event. Reports suggest that Apple plans to use the event to highlight its on-device AI capabilities as a competitive advantage, leveraging its 15 years of custom silicon expertise to showcase how AI systems can run directly on devices rather than relying heavily on the cloud.

The Cupertino-based company is expected to demonstrate how the chips designed for iPhones, Apple Watches, and Macs give it an edge in processing AI queries locally on devices. While cloud-based processing will still remain necessary for more complex queries, Apple is expected to position local inference as a privacy-preserving and cost-saving alternative to the massive data centre buildouts pursued by many of its rivals over the years.

More from Tech

As part of its agreement with Google, Apple is reportedly set to use a larger version of Google’s Gemini model to train a smaller distilled version capable of running locally on Apple hardware. Reports also suggest that Apple is actively exploring acquisitions to help advance its model-shrinking efforts.

Apple is also believed to have approved the use of Nvidia’s confidential computing technology within Google Cloud to handle processing for the larger Gemini-based model. The security feature encrypts data and AI models during processing, adding a modest performance cost while offering stronger privacy protections.

These new reports represent a departure from Apple’s original Apple Intelligence announcement, in which the company had stated that all cloud-bound queries would be handled exclusively through its own Private Cloud Compute infrastructure. Despite these changes, Apple is reportedly expected to retain the Private Cloud branding, according to people familiar with the matter cited by The Information.

With this new direction, Apple could soon run into practical limitations regarding how far it can push on-device AI processing. Google’s full Gemini model reportedly runs on trillions of parameters, and reports suggest that Apple has struggled to run it even on its own Private Cloud Compute infrastructure, which uses the same Apple Silicon chips found in Mac computers.

Apple Intelligence was first announced at WWDC 2024, but its rollout has reportedly been hampered by a lukewarm response to the initial feature set and delays surrounding the launch of a more personalised version of Siri. Apple is now expected to use WWDC 2026 to highlight these delayed features while also unveiling a new set of AI-focused capabilities.