蘋果將生成式 AI 整合至硬體並推出 Siri AI
Apple Integrates Generative AI into Hardware with Siri AI
Updated at: June 14, 2026 at 07:15 AM
2026年6月，蘋果宣布了一項重大的策略轉向，將生成式人工智慧直接整合到其硬體中。
In June 2026, Apple announced a major strategic pivot by integrating generative AI directly into its hardware.
這場變革的核心主角是「Siri AI」，這是對經典語音助理進行的一次徹底改造。
The star of this transformation is 'Siri AI,' a complete overhaul of the classic voice assistant.
與前代產品不同，Siri AI 作為一種對話代理運作，能夠理解螢幕情境並管理各應用程式間的個人數據。
Unlike its predecessor, Siri AI functions as a conversational agent that understands screen context and manages personal data across apps.
該系統構建於第三代蘋果基礎模型（Apple Foundation Models）之上——這是在與 Google 的 Gemini 合作下開發的——並同時兼顧了效能與隱私。
Built upon third-generation Apple Foundation Models—developed in partnership with Google’s Gemini—this new system prioritizes both power and privacy.
蘋果採用了一種混合處理模型，利用高效能的裝置端蘋果晶片（Apple Silicon）處理常規任務，並透過安全的「私有雲端運算（Private Cloud Compute）」處理複雜請求。
Apple employs a hybrid processing model, using high-performance on-device Apple Silicon for routine tasks and a secure 'Private Cloud Compute' for complex requests.
隱私仍是至關重要的核心；其架構確保個人資訊既不會被儲存，公司也無法存取。
Privacy remains paramount; the architecture ensures that personal information is neither stored nor accessible by the company.
除了語音之外，該系統還引入了「視覺智慧（Visual Intelligence）」，讓使用者能夠透過相機與環境互動，以及全系統通用的寫作工具以實現無縫編輯。
Beyond voice, the system introduces 'Visual Intelligence,' allowing users to interact with their environment via camera, and system-wide writing tools for seamless editing.
這些進階功能需要現代的蘋果晶片，例如 M3/M4 晶片或 iPhone 17 Pro。
These advanced capabilities require modern Apple Silicon, such as the M3/M4 chips or the iPhone 17 Pro.
隨著蘋果從「Core ML」轉向更強大的「Core AI」框架，這標誌著一個新時代的到來：生成式智慧將作為整個蘋果生態系統的基礎介面層，使科技變得比以往任何時候都更加直觀且緊密相連。
As Apple transitions from 'Core ML' to the more robust 'Core AI' framework, it signals a new era where generative intelligence serves as the foundational interface layer across the entire Apple ecosystem, making technology more intuitive and interconnected than ever before.
