At its 2026 Worldwide Developers Conference, Apple revealed five foundational AI models developed jointly with Google, underpinning advancements in Siri and various AI-powered features designed to run securely and privately across Macs, iPhones, and other Apple devices.
- Five foundational AI models co-developed by Apple and Google
- Enhanced Siri uses multimodal AI with improved context and privacy
- AI models run on-device and via Apple’s private cloud infrastructure
What happened
During its Worldwide Developers Conference in June 2026, Apple announced the launch of five new Apple Foundation Models, developed in collaboration with Google. These foundational AI models provide the technical basis—akin to roads rather than cars—for Apple's AI-driven software and features, particularly a revamped Siri that operates across iPhone, Mac, Watch, and AirPods. The models utilize a mixture of publicly available, licensed, and synthetic data for training, while respecting privacy standards by excluding users’ data from the training process.
The AI models are designed to run efficiently on-device, reducing reliance on cloud processing, although Apple also employs its private cloud infrastructure which uses Google Cloud and Nvidia GPUs. Apple ensures customer data privacy by not storing chat logs or personal information in the cloud. The new AI capabilities are integrated into macOS Golden Gate and upcoming iOS 27 features, with Siri upgraded to better interpret speech and images, and coordinate across various apps seamlessly.
Why it matters
This collaboration marks a substantial industry milestone as two major tech giants uniting on foundational AI technology could accelerate the adoption of advanced, privacy-centric AI solutions. Apple’s emphasis on privacy, on-device processing, and respect for data rights reflects its strategic positioning against competitors like Google and Anthropic, which focus heavily on cloud-based or large-scale generative AI models. It signals a shift toward embedding AI more seamlessly and securely into everyday consumer technology.
Apple’s approach prioritizes a natural, trustworthy, and discreet AI experience integrated deeply into the operating system and user workflows. Analysts suggest that if Apple's AI implementations succeed in feeling invisible yet powerful to users, it could strengthen its ecosystem significantly and redefine mainstream user expectations around AI usefulness, privacy, and accessibility, especially as concerns over data privacy and security continue to rise globally.
What to watch next
Attention will focus on how developers leverage Apple’s Core AI tools to build new applications that take advantage of these foundational models, particularly in sectors requiring high privacy standards. Observers will also monitor adoption rates of the updated Siri and related AI functionalities across Apple's ecosystem and how well the on-device AI performs compared to cloud-reliant competitors.
Further updates in upcoming software releases like iOS 27 and macOS Golden Gate will reveal how deeply these AI models integrate across device capabilities and workflows. Industry watchers will also evaluate whether the partnership with Google around cloud infrastructure scales effectively without compromising on Apple’s hallmark privacy commitments, marking Apple’s progress toward making AI a seamless and secure part of the everyday user experience.