Apple will rely on Google's fleet of Nvidia chips to power its overhauled version of Siri when it launches in September, according to a new report from The Information.
Last week, the outlet reported that Apple plans to highlight the on-device AI capabilities of its devices at WWDC next week, but queries that require cloud-based processing will still fall back on one of Google's large Gemini models, as per an agreement between the two companies.
Today's report adds some specificity to the planned cloud setup by revealing that Apple will tap into Google's fleet of Nvidia Blackwell B200 data center chips, where user data will be encrypted using Nvidia's hardware-based confidential compute feature. Introduced in 2024 as the successor to Hopper, Blackwell chips are designed primarily for large language models, and can dramatically speed up AI training and inference compared to the previous generation.
The report notes that the arrangement diverges from Apple's usual strategy of "attempting to control all the critical ingredients to its products." It also adds that it's unclear how Apple's previously launched server system, called Private Cloud Compute, will fit into the upcoming Siri launch.
Private Cloud Compute runs on Apple's Mac-series chips and was announced two years ago as a way to offer cloud-based computing in a more private and secure fashion. Apple reportedly tried to get a modified version of Gemini working on its in-house server system, but found that it ran too slowly. The publication's previous report said Apple will likely retain the Private Cloud Compute branding despite the change.
Apple Intelligence was unveiled at WWDC 2024, but its rollout has been overshadowed by a lukewarm reception to its initial features and ongoing delays to the more personalized version of Siri. WWDC 2026 begins on June 8, when Apple is expected to reset the narrative by revisiting those delayed features and introducing new AI capabilities.
