Why Apple and WeChat’s AI Rollouts Are Slower Than Expected

01 – Apple announced “Apple Intelligence” at WWDC 2023 with fanfare, but the first release suffered from limited functionality, delayed Siri updates, and performance‑heavy on‑device models that only work on newer iPhone models, making many features feel like a showcase rather than a usable product.

02 – WeChat has added AI‑related functions such as AI reading for public accounts, AI‑generated replies, and integration of large models like DeepSeek, yet these are merely entry points that are not deeply woven into the core social and messaging architecture, resulting in a fragmented user experience.

03 – Privacy is a core differentiator: Apple markets privacy as a selling point, while WeChat’s business model relies on data security; both companies therefore limit the amount of user data that can be fed into large‑scale model training, slowing AI advancement.

04 – Apple’s “three‑horse carriage” approach—on‑device models, private‑cloud deployment with end‑to‑end encryption, and user‑confirmed queries to external services—protects privacy but incurs higher power consumption and restricts model capabilities to newer hardware.

05 – WeChat’s AI remains superficial, with suggestions that future “active intelligence” could automate likes, reminders, and work‑group replies, but such features would demand extensive context handling and raise trust concerns.

06 – The concept of a WeChat‑specific Agent AI that leverages the platform’s social graph, content ecosystem, and millions of mini‑programs could provide a differentiated, mobile‑first AI experience, yet its development is still in early stages with no concrete timeline.

07 – Both Apple and WeChat adopt an “application‑first” roadmap, focusing on practical features rather than pushing the limits of AI research, which, combined with privacy constraints, explains their slower progress.

08 – The article concludes that privacy, data‑security, and pragmatic product strategy are the main reasons Apple and WeChat’s AI adoption is gradual, and that future breakthroughs will require balancing these factors with more ambitious model capabilities.