What did John Ternus say about Apple's AI strategy?
Apple CEO John Ternus used his first extended interview since taking the job to describe an AI strategy built around products rather than chatbots. WIRED published Steven Levy's interview on October 30, 2026, two days after Apple's touchscreen MacBook Pro event in Los Angeles.
Levy asked Ternus about two competing views of Apple's AI record. One holds that Siri fell behind after repeated delays. The other holds that Apple avoided the cost of building a frontier model. Ternus did not choose either framing. "Everyone has an opinion—I really like where we're at," he said.
The line that travelled was shorter. "We don't want to ship AI for AI's sake," Ternus said. He added that Apple is "absolutely going to leverage AI" in its products. The interview took place after Apple showed its new touchscreen laptop that week.
He then described what success should look like. "When we're doing it right, people won't have to think about whether it was AI that made this happen or more traditional software," he said. "What matters is, did we build something that they really love?"
TechCrunch and TNW both covered the interview on October 31. TNW noted that Ternus would not say whether he backs AI regulation, and instead framed the issue around security. That omission matters while governments write rules for models and assistants.
The interview offered no product roadmap, no shipment dates for future AI features and no AI revenue figures. It was a statement of philosophy from a CEO who had run Apple for less than two months.
That makes the timing important. Apple reports fiscal fourth-quarter results on November 2, and Ternus will face investors for the first time as chief executive. The interview reads as an opening argument for how he wants that conversation to go.
The scale of Apple's customer base is why the comment matters beyond Cupertino. The company says its installed base of active devices reached an all-time high in the June quarter. Its choice to make AI ambient rather than a separate app shapes how a large share of consumers first meet the technology.
Why is Apple putting AI inside products instead of selling it?
Apple's answer is already visible in its software. On September 14, 2026, Apple released the next generation of Apple Intelligence, including Siri AI, as part of its 2027 software updates. The assistant shipped in beta in English, with five more languages promised for the following month.
Apple's description stresses integration over novelty. Siri AI can use what is on the screen, search a user's messages and photos, and take actions across apps. Apple also added AI tools to Photos, Safari, Image Playground and other built-in apps.
The features show what Apple means by integration. Siri AI can draw on personal context, understand what is on screen and carry out actions inside apps. Apple also added generative editing in Photos, new Safari browsing tools and an Image Playground that can create photorealistic images.
The models are not entirely Apple's. Apple says its Apple Foundation Models were "custom-built in collaboration with Google and its Gemini models." They run on the device and on servers through Private Cloud Compute.
Apple uses that architecture to make a privacy argument. The company says Private Cloud Compute extends the privacy and security of the iPhone into the cloud, so requests that need more compute can be handled with iPhone-level protections. That claim is central to Apple's pitch to cautious customers.
The rollout is still partial. Siri AI launched in beta in English, with French, Japanese, Korean, Portuguese and Spanish promised for the next month. Apple frames the features as part of its 2027 software cycle, so many users will meet them through updates rather than a new device.
That arrangement explains the business logic. Apple sells hardware and services, so AI earns its keep by improving devices people already buy. WIRED describes Apple as a $4 trillion company with a $100 billion services business.
The approach has clear trade-offs. Apple controls the integration and the privacy story, but it depends on a partner for part of the underlying intelligence. It also competes with companies that sell AI as a standalone product, where progress is easier to see and market.
Ternus rejected the idea that AI agents will make apps and products irrelevant. "That could make life really easy, I suppose," he said, but he argued that "the creation of products is a deeply human and personal thing."
For users, the bet is that AI becomes ordinary. For Apple, the bet is that trust and integration matter more than leading a benchmark. The company is asking customers to judge the result, not the model label.
How does Apple's AI spending compare with Microsoft, Google, Amazon and Meta?
The contrast with Apple's largest peers is stark. BBC reported on October 30 that Meta, Alphabet, Oracle, Amazon and Microsoft have committed $800 billion between them for 2026.
The same report said Alphabet expects to spend up to $205 billion this year and Amazon has committed $220 billion. Much of that money is going into AI data centres, network equipment and servers.
Investors have started to question the returns. The Nasdaq ended October with its worst monthly slide since March 2025, and Intel shares fell 28% in a day after the chipmaker raised spending despite weaker PC demand.
Apple is not in that group. Its AI costs are spread across chip design, partner models and cloud processing rather than a single fleet of training clusters.
That distance is a strategic choice, not an oversight. Apple designs its own silicon and relies on partners for part of the model stack. The arrangement keeps Apple out of the spending race that has rattled investors, but it also means Apple does not control the pace of every improvement.
That choice lowers financial risk but raises dependence risk. If Gemini-based features improve quickly, Apple benefits without paying for the whole training run. If Google's roadmap stalls, parts of Apple's assistant would lose a key input.
Apple's payoff will also be harder to measure. It will show up indirectly, in device upgrades and services revenue, not in a separate AI line item. That makes the strategy difficult to judge quarter by quarter.
The company does have scale on its side. Apple reported $109.4 billion in revenue for its June quarter, up 16% from a year earlier, with record June-quarter revenue from iPhone, Mac and Services. AI has to support that business, not replace it.
Who is John Ternus and what has changed at Apple?
Ternus is a 25-year Apple veteran who became CEO on September 1, 2026. Apple's board approved the succession unanimously, and Tim Cook moved to executive chairman after leading the company through the iPhone era.
Apple presented the handover as a promotion of its product culture. Cook said Ternus has "the mind of an engineer, the soul of an innovator," language that framed the change as continuity rather than a reset.
Cook's new role is not ceremonial. Apple said he would assist with "engaging with policymakers around the world," a job that includes the company's most sensitive political relationships.
Ternus described the working arrangement in the interview. "Tim is available to help wherever I ask him, and it's great to have him in that role," he said. WIRED reported that Ternus has not yet met President Donald Trump.
The engineering organization also changed. Apple merged its hardware engineering and technology groups under Johny Srouji, who reports to Ternus. Srouji called Ternus "a product guy at heart" and said Apple's culture would not change.
Silicon sits at the center of that structure. WIRED describes Srouji as the architect of Apple's custom silicon, an initiative that has driven much of the company's hardware work over the past decade. Putting hardware and technology under one leader gives Ternus a single executive for the chips, devices and systems that AI features depend on.
Marketing chief Greg Joswiak made a similar point. "John innately has a great sense for great products, and that's what Apple is," he said. "You can't teach that."
The first product test came this week. Apple announced its first touchscreen MacBook Pro on October 28, starting at $2,899 and shipping November 6. The launch restored the illuminated Apple logo, a small detail that fits Ternus's hardware-first image.
Ternus framed himself as a continuation rather than a rupture. He said he had only "a handful of interactions" with Steve Jobs but felt like "a student of Steve's teachings." He also said he had not really thought about which non-Apple CEOs he admires.
That is the central tension of his early tenure. He is selling continuity at a moment when AI is forcing every large technology company to change, and his main credential is product judgment rather than an AI research background.
What did Ternus leave unanswered about regulation, politics and agents?
The clearest omission was regulation. Levy asked whether Ternus favors AI regulation, and Ternus answered by talking about cybersecurity. "We all read the stories and think about it from a cybersecurity standpoint—clearly these models are very capable of doing things," he said.
He added that Apple's focus is on secure, safe products and protecting user information. He did not endorse new rules, oppose them or set out conditions for supporting them.
Politics was another gap. WIRED reported that Ternus has yet to meet Trump, while Cook remains involved in policy. Apple's April announcement said Cook would help with policymakers, which gives the new CEO cover on difficult meetings.
Agents drew a more direct answer. Ternus doubted that AI agents would replace the apps and products people use. He said product creation is "a deeply human and personal thing," and that Apple's job is to make tools people love.
Those non-answers carry risk. Apple operates in markets with active AI rules, including the European Union, and its assistant depends on partners whose models face scrutiny. A CEO who does not state a position has less influence when rules are drafted.
They also carry a political benefit. A leader who avoids committing to specific rules keeps room to negotiate later, especially while the technology and its regulation are still moving.
The agent question has commercial stakes too. If AI agents became the main way people interact with software, the App Store's role as a distribution gate would shrink. Ternus's defence of products is also a defence of the platform model that supports Apple's services revenue.
The interview's limits are worth noting. It was one conversation, conducted around a product launch, and it produced no commitments on model transparency, safety testing or the next Siri release. Readers should treat it as a signal of priorities, not a policy document.
What should investors and users watch next?
Apple reports fiscal fourth-quarter results on November 2 at 2 p.m. PT. It will be Ternus's first earnings call as CEO, and the first formal test of how he explains the AI strategy to shareholders.
The call will show whether Apple's product cycle is holding up. It will also give executives a stage to explain how AI features feed hardware and services sales, rather than appearing as a cost with no clear return.
Apple's June quarter gives a baseline. Revenue was $109.4 billion, up 16%, and gross margin was 50.1%. Investors will compare the September quarter with those records and listen for guidance on the December quarter, including signs that AI features are lifting upgrade rates or services spending.
Three product questions matter. First, whether Siri AI moves beyond its English beta and reaches more users. Second, whether the touchscreen MacBook Pro broadens Apple's upgrade pool. Third, whether Apple can sell AI-heavy devices while the memory chip shortage raises costs, a pressure Ternus has already called a crisis.
Apple's language rollout is another gauge. Siri AI began in English only, and Apple said French, Japanese, Korean, Portuguese and Spanish would follow within a month. The speed and quality of that expansion will shape how quickly the assistant reaches Apple's global customer base.
The Google relationship is the wildcard. Apple's official release credits Gemini models, and any change in that partnership would affect both capability and cost. Ternus did not discuss the terms in the interview.
The regulatory question is just as important. Apple's AI rollout varies by market, and any delay in a large market changes the return on features built for the whole product line.
The strategic question is larger. If Apple can make AI feel like a normal part of the iPhone and Mac, it can keep its pricing power without buying frontier-model scale. If rivals make standalone assistants the main interface, Apple's integration advantage weakens.
Ternus's answer for now is that the product comes first and the AI should disappear into it. The next year of shipments, software updates and regulatory decisions will test whether that is confidence or complacency.