Apple deepens reliance on Google, hands Siri over to Google Cloud

Apple is seeking to host the next-generation version of its voice assistant Siri on Google Cloud, reflecting the deep-rooted dilemma resulting from the tech giant’s long-term underinvestment in cloud computing, and marking an increased reliance on external infrastructure. According to The Information’s report on Tuesday, at Apple’s request, Google has begun a feasibility study on deploying dedicated servers inside its data centers to run the new Siri powered by Gemini. Sources reveal that the arrangement must also meet Apple’s privacy standards. Apple has previously relied on Google Cloud for online storage and internal AI model training. If this hosting plan comes to fruition, the cloud computing partnership between the two companies will enter a new phase. Behind this move is the severely low utilization rate of Apple’s own cloud infrastructure. Former employees reveal that Apple’s private cloud computing system designed for AI has an average utilization rate of only 10%, with some servers even still stored in warehouses and not installed. Meanwhile, Apple’s progress in AI continues to lag behind expectations. The more conversational new Siri has yet to be launched, and the market response to its released AI functions has been tepid. Financial Orientation Suppresses Strategic Investment in Cloud Computing Apple’s predicament in cloud computing can be traced back decades, rooted in the company’s finance team long viewing cloud computing as a cost burden rather than a strategic asset. According to about a dozen former Apple executives and engineers, Apple’s finance department has consistently suppressed large-scale investment in its own cloud infrastructure, preferring to rent computing power from external suppliers such as Google and Amazon Web Services (AWS) to maintain flexible control over infrastructure costs. This orientation has caused ongoing talent loss in cloud computing. Mike Abbott, who joined Apple in 2019 to lead cloud infrastructure integration, departed in 2023, and many team members soon followed him to General Motors, further weakening Apple’s internal cloud capabilities. In terms of capital expenditure, Apple’s gap with peers is especially stark. Tech companies like Meta, Microsoft, Google, and Amazon have in recent years invested heavily in data centers to cope with explosive growth in AI computing needs, while Apple has largely stayed on the sidelines. Frequent Issues with Private Cloud System Restrict AI Implementation Apple’s private cloud computing system, released alongside Apple Intelligence in June 2024, was in fact incomplete at launch, delayed about six months from its original schedule, and will not officially go online until the end of 2024. Former employees point out that the system’s performance in Apple’s own data centers is unsatisfactory. First, software updates for private cloud computing servers take far longer than for other server types; second, servers equipped with Apple’s custom chips were not designed for AI workloads and have clear limitations when running large models like Google’s Gemini. Igor Naverniouk, a former Apple engineer who worked on next-generation Siri and left in December last year, highlighted the fundamental difference in engineering culture between Apple and Google: “At Google, most resources are centrally managed, and everyone shares the same supercomputer. At Apple, technology choices are fragmented.” He previously worked on Google’s infrastructure team for more than ten years. Google Partnership: From Forbidden Zone to Core Dependency Apple’s collaboration with Google in cloud computing hasn’t always been smooth. For years, Apple, citing privacy concerns, strictly prohibited AI engineers from using Google Cloud—software chief responsible for privacy, Craig Federighi, repeatedly vetoed proposals to use Google Cloud for AI computing. The turning point came in 2023. Google upgraded its security systems to meet Apple’s privacy requirements; Apple then began integrating Google Cloud into its AI infrastructure and adopted Google’s self-developed TPU chips—Apple’s calculations indicated that their operating costs were significantly lower than Nvidia’s comparable chips. In January this year, Apple announced an agreement with Google to bring the Gemini model into Apple products, further confirming a deep AI tie-up between the two companies. If Google ultimately hosts the new Siri in the cloud, Apple’s reliance on this competitor will extend from the model layer to the infrastructure layer, becoming a more comprehensive strategic dependency. New Siri May Become a Pivotal Opportunity Despite the low utilization rate of its private cloud computing system, Apple has stated it will launch the new Siri this year. Once widely adopted by users, AI computing demands may quickly soar, posing a severe test for whether the existing infrastructure can cope. The Information, citing insiders, said discussions between Apple and Google about hosting the new Siri are partly to lay groundwork early in anticipation of surging compute demand after the new Siri goes live. For investors, these developments reveal Apple’s structural weaknesses in the AI race: as competitors continue to ramp up their own compute infrastructure, Apple’s cloud capability construction remains hampered by financial culture and historic inertia. The gap between the execution pace of its AI strategy and market expectations will be hard to close in the short term. Risk Warning and Disclaimer The market carries risks and investment should be done cautiously. This article does not constitute personal investment advice and does not account for any individual user's unique investment objectives, financial situation, or needs. Users should consider whether any opinions, views, or conclusions herein are suitable to their specific circumstances. Investing based on this article is at your own risk.