Apple reportedly plans to use next month’s Worldwide Developers Conference (WWDC) to highlight its on-device AI capabilities as a competitive advantage, drawing on 15 years of custom silicon expertise to make the case for running AI models locally rather than in the cloud.
People familiar with Apple’s plans speak to The information say the company should show how chips designed for iPhones, Apple Watches and Macs give it an edge in processing AI queries directly on devices. Although cloud-based processing will still be necessary for complex queries, Apple will position local inference as a cost-effective, privacy-preserving alternative to the massive data center buildouts that its competitors have pursued.
As part of its deal with Google, Apple is apparently willing to use a large version of Google’s Gemini model to form a smaller, distilled version capable of running locally on Apple hardware. Apple is also reportedly looking for acquisitions to help advance its model reduction work, with one company it’s reportedly considering being Liquid AI, a Massachusetts startup focused on running AI locally on devices.
Some queries will still require cloud processing. Apple has reportedly approved the use of Nvidia’s confidential computing technology within Google Cloud to handle processing for the broader Gemini-based model. The security feature encrypts data and AI models during processing, which adds a modest performance cost but provides stronger privacy protection.
The deal represents a notable change from Apple Intelligence’s initial announcement, in which the company said all cloud-related queries would be handled exclusively by its own Private Cloud Compute infrastructure running on Apple silicon. Apple will likely keep the Private Cloud Compute brand despite the change, people familiar with the partnership said. The information.
There are also hardware limits to how far Apple can push processing onto the device. Google’s comprehensive Gemini model draws on billions of parameters, and The information claims that Apple struggled to run it 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 the rollout was hampered by a lukewarm response to initial features and a prolonged delay in the more personal version of Siri. Apple is now expected to use WWDC 2026, which takes place on June 8, to reframe the narrative, reintroduce delayed features and launch new ones.
