When the world's most valuable technology company repeatedly promotes "Liquid Glass" design at length, investors realize that what they have been waiting for is not an AI revolution, but a signal of Apple's waning innovation. The global developer conference (WWDC) on June 10, Beijing time, is destined to become one of the most disappointing developer conferences for Apple in recent years. This tech giant, with a market value exceeding $3 trillion, delivered a lackluster performance at a critical juncture in the AI race. Aside from the redesign of the "Liquid Glass" interface, there were almost no groundbreaking innovations that could shake up the market landscape. A research report from Huatai Securities points out that Apple is deeply trapped in the "innovator's dilemma," facing threefold challenges of insufficient edge capabilities, declining R&D efficiency, and privacy protection constraints, which may provide an opportunity window for Chinese companies like Xiaomi and Lenovo to close the gap. Due to insufficient foundational capabilities such as chips, killer AI applications on edge hardware like smartphones have yet to emerge; As an industry leader, Apple plans to invest as much as $31.4 billion in R&D for FY24, yet still fails to close the gap with competitors like Google in AI, with a significant decline in R&D conversion efficiency; Privacy protection is gradually becoming a factor that restricts AI progress, and the strategic positioning of various companies may affect their AI development speed. Apple's strict privacy protection for users is becoming a significant obstacle to its AI development. UI Refresh Fails to Conceal Weakness in AI Innovation Although Apple has launched the "Liquid Glass" design language covering iOS 26, iPadOS 26, macOS 26, and watchOS 26, it has only achieved minor upgrades in AI functionality. Apple claims to have showcased "the largest software upgrade in years" at WWDC, changing its operating system naming convention from serial numbers to using years (e.g., iOS 26 corresponds to 2026), and introduced the "Liquid Glass" design language, achieving visual optimizations such as adaptive lock screen wallpapers and full-screen browsing in Safari. In terms of functionality, the Phone app has added call screening and music recognition waiting features, CarPlay integrates a weather plugin, and Messages supports chat background customization. However, these improvements have been evaluated by the Huatai report as "many of which have already been implemented on Android devices." The most notable AI advancements from Apple are merely minor upgrades: Apple Intelligence has added real-time voice translation, fitness partner voice encouragement, visual search (screenshot recognition of products/events), and opened up a foundational model framework to developers. This falls far short of the disruptive AI experience the market expected. An earlier article from Wall Street Watch mentioned that at last year's global developer conference (WWDC), Apple heavily promoted how Apple Intelligence would enhance Siri's capabilities Recently, the media predicted that this year's WWDC AI updates would be limited, and the upgrade of Siri would still be absent. Indeed, at this conference, Apple's Senior Vice President of Software Engineering, Craig Federighi, admitted that these AI features for Siri would not be launched in the short term. Huatai Securities stated that Apple is deeply trapped in the "innovator's dilemma." Dilemma 1: Insufficient Edge Computing Capabilities Leading to Delayed Killer AI Applications Huatai's report lists "insufficient edge computing capabilities" as Apple's primary dilemma. It believes that the main reason why killer applications for edge AI have not emerged is that the basic capabilities of current devices such as smartphones and PCs in terms of chip computing power, accessibility of user data, and operating system capabilities are still immature, limiting the further development of AI. Over the past year, the team has repeatedly observed similar issues across various ecosystems, including Android phones and Windows PCs, which may continue to extend the replacement cycle for PCs and smartphones, ultimately affecting overall sales. Dilemma 2: R&D Investment Snowballing, Innovation Output Stalling Apple's R&D funding is trapped in the "diminishing marginal returns" curse. According to its fiscal year 2024 report, the company spent a whopping $31.4 billion (approximately RMB 227 billion) on R&D, a year-on-year increase of 5%, accounting for 8% of total revenue. This is equivalent to burning $8.6 million every day. However, the massive investment has not translated into innovative breakthroughs. Huatai Securities wrote: After years of investment, the company halted the development of the electric vehicle "Titan Project" in 2024, and there have been multiple delays in the upgrade timeline for AI features, failing to narrow the gap with OpenAI, Google, and others in AI. Apple's R&D conversion efficiency has shown a significant decline, indicating that industry leaders, upon entering a mature business phase, have seen their innovation capabilities decline in order to maintain their existing profit models. In contrast, Chinese tech companies are demonstrating higher cost-effective R&D efficiency. Taking Xiaomi as an example, its R&D investment in 2024 was RMB 24.1 billion (about 10% of Apple's), yet it has made substantial progress in the electric vehicle and self-developed chip fields. The Huatai report believes that latecomers like Xiaomi and Lenovo are leveraging their late-mover advantage to close the gap with Apple. Dilemma 3: The Dilemma of Privacy Protection and AI Evolution Apple's pride in privacy protection is becoming a "double-edged sword" in the AI era. To fulfill its data security commitments, Apple has built a private cloud architecture and refuses to use user data on a large scale for model training. Companies like Google and Meta iterate their models through massive user data, while Apple's strict limitations result in insufficient training data for its large models. The Huatai report points out that a relatively relaxed level of privacy protection would be more beneficial for the advancement of AI technology In the AI era, Apple's relatively strict privacy protection for users is becoming a significant obstacle to its development of AI. In terms of privacy, although companies like Google, Amazon, Meta, Microsoft, as well as Huawei, Xiaomi, and ByteDance will encounter similar issues, their relatively looser privacy protection levels will be more conducive to advancements in AI technology. This contradiction is particularly prominent in cross-device collaboration scenarios. For example, the "Visual Intelligence" feature promoted by Apple only supports local screenshot recognition and cannot call upon a cloud image database for more precise matching like Google Lens. Apple has maintained a relatively conservative attitude towards public cloud AI businesses, which require significant investment in computing power. The lack of investment in data and computing power has also led to Apple's cloud large model capabilities being significantly behind competitors like OpenAI and Google. Apple ultimately hopes to turn the tide at this fall's iPhone launch event. According to the event, the new generation of iPhone will deeply integrate Apple Intelligence, but specific technical details have not yet been disclosed