Apple's new CEO Ternus is dissatisfied with AI progress and demands the software team move beyond the Siri patch-up phase

As of the close on November 9, Eastern Time, Apple had risen about 29% year to date, making it the only member to significantly outperform the S&P 500. By contrast, shares of giants such as Google, Meta, and Nvidia, which are heavily invested in AI infrastructure, all fell to varying degrees, while Tesla dropped more than 20%.

Stripping away the surface of stock prices, Apple's resilience does not stem from leading AI technology, but from its "asset-light" financial foundation. While other giants are dragged into a cash-flow quagmire by hundreds of billions in capital expenditure, Apple, with extremely low capital expenditure and abundant free cash flow, has secured its position as a safe-haven asset.

Why is Apple the most resilient?

According to a compilation by a reporter from the Science and Technology Innovation Board Daily, as of the close on November 9, Eastern Time, Apple had risen about 29% year to date, making it the only member among the Magnificent Seven to significantly outperform the S&P 500. Over the same period, Microsoft rose about 12%, Amazon rose about 15%, and Nvidia fell about 6%; Meta and Google fell about 3% and 8% respectively, while Tesla dropped about 20%.

It can be said that Apple is currently the most resilient of the seven major U.S. technology giants. To understand Apple's resilience, one first needs to see clearly what the other technology giants experienced in 2026.

In the third quarter of 2026, the combined capital expenditure of Amazon, Alphabet (Google's parent company), Microsoft, and Meta reached $201.9 billion, and all four companies' full-year capital expenditure plans exceed $100 billion. Over the past three and a half years, these four companies have invested more than $1 trillion cumulatively in AI infrastructure, almost all of it directed at data centers, GPU clusters, and computing networks.

The cost of becoming asset-heavy has already emerged. In Q3 2026, the free cash flow of Alphabet and Meta turned negative; although Microsoft's free cash flow remained positive, it fell 27% year on year; Amazon's free cash flow over the past 12 months was negative $42.4 billion.

The asset-heavy shift in capital expenditure has changed not only cash-flow figures, but also the framework by which the market prices such companies. As depreciation begins to continuously erode profits and free cash flow turns from positive to negative, investors need to keep answering the same question: when will this capital expenditure translate into returns? The market's pricing framework for such companies is also shifting from "high-margin software companies" to "infrastructure companies that require continuous investment to maintain operations."

On the other hand, Nvidia, at the other end of the AI chain, has very low capital expenditure of its own—only $6 billion in FY2026—yet its shares fell about 6% year to date. The reason is that, as the "shovel seller," Nvidia's performance is highly dependent on purchases by those four giants. When buyers face doubts about peak capital expenditure due to deteriorating free cash flow, the sustainability of Nvidia's orders is naturally called into question.

Tesla faces a different predicament. It is not involved in the arms race over large-model data centers, but its Q3 R&D expenses surged 69% to $2.76 billion, while operating profit fell 56% year on year to $718 million and its operating margin dropped to 2.6%, compounded by declining deliveries in its core automotive business and pressure on energy business revenue. Tesla's AI narrative remains grand, but in the short term it is difficult to fill the hole left by declining profits in its core business.

By contrast, Apple has neither made heavy asset investments as a buyer nor engaged in cyclical speculation as a seller. Its capital expenditure for the full fiscal year 2026 was only $9.04 billion, down 29% from $12.7 billion in the previous fiscal year. Financial reports show that Apple's full-year revenue was $478.8 billion, up 15% year on year; net profit was $132.2 billion, up 18% year on year; operating cash flow was $145.097 billion, up about 30% year on year; and free cash flow was $136 billion, up about 38% year on year.

English text is being prepared.

Zhang Yirui pointed out that the market is now strictly scrutinizing ROI and free cash flow for AI giants. "The pricing anchor for other giants lies in the cloud elasticity converted from high AI CapEx, while Apple's pricing anchor lies in the installed base of more than 2 billion active devices, high-margin service cash flow, and highly certain share buybacks."

In his view, a restrained AI pace is key to Apple's safe-haven status. "Apple did not engage in aggressive heavy-asset arms races in general large models and ultra-large-scale GPU clusters in the early stage, and instead avoided the current market concerns about excess capital expenditure and depreciation eroding profits," Zhang Yirui told a reporter from the Science and Technology Innovation Board Daily.

According to media reports, Apple's new CEO Ternus recently held a closed-door meeting with software chief Craig Federighi. Ternus expressed dissatisfaction with the company's current progress in AI and demanded that it should not be satisfied with patching up the existing Siri, but should formulate a long-term plan.

It is reported that at the closed-door meeting, Ternus asked the software team to "quickly move beyond the stage of merely making necessary improvements to Siri." Apple is considering giving the voice assistant Siri stronger autonomous action capabilities, and could launch a new version of Siri with agentic capabilities as early as 2027.

It is worth noting that Ternus had previously always maintained a cautious stance on artificial intelligence. In an interview with Wired magazine on October 30, he said, "We don't want to ship AI for AI's sake." Apple will embrace AI, but does not believe in putting it into products simply because the technology is trending.

Why is Apple "unable to sit still"? Because the competitive paradigm is shifting. OpenAI, Meta, and xAI have pushed the focus of competition from "a better chat assistant" to "goal-driven agents," in which systems continuously track goals in the background and autonomously decide the next action, rather than waiting for users to issue instructions one at a time.

It is not that Apple has made no progress. Siri AI, which Apple has launched in the U.S. market, already has cross-app operation, screen awareness, and multi-step task execution capabilities, and is no longer a simple single-step instruction tool. But it still remains in a "request-response" mode: the user issues an instruction, Siri executes it, and then waits.

Among phone makers, Huawei, Xiaomi, Honor, and others in the Chinese market have already gone through multiple rounds of iteration around on-device large models and system-level agents. However, Apple's China-market Apple Intelligence did not officially announce until October 31 that it would integrate Alibaba's Tongyi Qianwen, with the first batch of about 18 features to be pushed in December with iOS 27.3. These cover basic capabilities such as writing tools, real-time translation, photo erasure, and notification summaries, but this update does not include Siri AI or other next-generation Apple intelligence features announced at WWDC 2026.

Internal pacing is also under pressure. According to media reports, Federighi's software department is accustomed to planning on a 12-to-18-month cycle, while the hardware department lays out plans years in advance; this mismatch easily makes first releases more conservative. Apple has already adjusted the initial pace of iOS/iPadOS 28 and macOS 28 from "four weeks of coding, four weeks of integration" to "six weeks of coding, two weeks of integration."

At present, the AI investment of tech giants is mainly capital expenditure—buying GPUs and building data centers. Financially, this money is not counted as a current cost but becomes "fixed assets," which are gradually amortized over the next 5 to 10 years through annual depreciation. The giants' free cash flow has therefore turned sharply negative, and in the future they face the risk of massive depreciation suppressing profits.

Apple's increased spending, by contrast, mainly adds R&D expenses, chip BOM costs, and the like. R&D expenses are current operating expenditures that directly reduce current profit and do not create a subsequent depreciation burden, and this kind of marginal cost can be shifted to end consumers through supply chain management and tiered product pricing.

English text is being prepared.

Zhang Yirui further pointed out: "As the gatekeeper of hardware and systems, on-device AI directly raises the threshold for upgrading devices and hardware ASP, while firmly locking in the super entrance to billions of high-net-worth users. No matter which external large model wins, anyone connecting to Apple's ecosystem must accept its platform revenue sharing and rules, so the commercial loop is relatively stable."

Apple's financial reports show that in fiscal 2026 Apple's R&D expenses were $45.765 billion, up about 32% year on year; capital expenditure was only $9.041 billion, down about 29% year on year. Chief Financial Officer Parekh said at the earnings call that the company would "significantly increase overall AI investment," with investment mainly concentrated in R&D. The more than fivefold gap between R&D expenses and capital expenditure also means that Apple's AI catch-up is far "lighter" on the balance sheet than that of other giants.

But "light" does not mean "cost-free." R&D expenses are real current expenditures, and every dollar Apple spends catching up in AI capability will directly hit operating margin, rather than being spread over many future years like depreciation. At the same time, rising memory prices are squeezing gross margin from another side. Ternus called memory prices "unprecedented," and Apple has already successively raised prices for Mac, iPad, and iPhone to hedge.

English text is being prepared.

English text is being prepared.