Apple Hires AI Veteran Amar Subramanya as New VP of AI Strategy

Apple has brought in veteran AI researcher Amar Subramanya as its new Vice President of AI, marking one of the most significant leadership changes in the company’s recent history. He steps into the role as John Giannandrea, Apple’s long-time senior vice president for Machine Learning and AI Strategy, prepares to retire in 2026 after a transition period in which he will serve as an advisor.

The move underlines how central artificial intelligence has become to Apple’s strategy. The company has faced sustained criticism for moving more slowly than rivals in visible, consumer-facing AI features such as chatbots and advanced voice assistants. At the same time, it has been rebuilding its internal AI organization around “Apple Intelligence,” a suite of models and tools designed to work tightly with Apple’s hardware and operating systems.

By appointing a leader who has worked at the heart of AI efforts at both Google and Microsoft, Apple is signaling that it wants to accelerate this shift and reduce the gap with competitors.

Who Is Amar Subramanya?

Amar Subramanya is described by Apple as a “renowned AI researcher,” and his career trajectory supports that description. He has spent decades in applied artificial intelligence, particularly in building large-scale systems that directly support consumer products.

His background includes two major chapters:

• Sixteen years at Google: Subramanya spent the bulk of his career at Google, eventually becoming head of engineering for the Gemini assistant and related AI systems. Gemini is Google’s family of large language models and AI assistants that sit at the core of many of its newer search, chat, and productivity experiences. Working on Gemini meant dealing with high-stakes issues reliability at massive scale, safety controls, and integration into products used by hundreds of millions of people.
• Recent role at Microsoft: Earlier in 2025, he left Google to join Microsoft as Corporate Vice President of AI. In that role, he was part of the leadership pushing forward Microsoft’s AI roadmap, including its Copilot experiences integrated into Windows and Office. That mix of experience—deep technical leadership at Google and executive-level responsibility at Microsoft—now comes with him to Apple.

For Apple, this profile is particularly useful. Subramanya understands both the research side of AI (training and refining large models) and the product side (shipping those models in ways that are safe, responsive, and aligned with a company’s brand and privacy commitments).

What Will Subramanya Lead at Apple?

At Apple, Subramanya will report to Craig Federighi, Apple’s Senior Vice President of Software Engineering. That reporting line is important. It puts the AI function directly under the executive responsible for iOS, macOS, and the broader software ecosystem, rather than as a parallel silo.

Apple says Subramanya will oversee several critical areas:

• Apple Foundation Models: These are the large models—language, vision, and multimodal—that power “Apple Intelligence” features such as rewriting text, summarizing content, generating images, and supporting advanced Siri capabilities. His leadership here will shape how powerful, efficient, and private these models become on Apple devices.
  
• Machine Learning Research: This includes long-term research into new architectures, training methods, and on-device optimization, especially for Apple silicon. Apple has historically focused on AI that can run locally on the device where possible, to reduce latency and protect user data.
  
• AI Safety and Evaluation: As generative AI systems become more capable, safety testing and evaluation are essential. It involves checking for bias, harmful outputs, privacy risks, and misuse. Subramanya’s team will be responsible for designing frameworks that test models before they reach hundreds of millions of users.
  

Taken together, these responsibilities make the new VP of AI one of the pivotal roles inside Apple. He will not only guide research, but also help decide how far and how fast Apple exposes generative AI to everyday users.

John Giannandrea’s Tenure and Planned Retirement

John Giannandrea joined Apple in 2018 after a high-profile career at Google, where he led search and artificial intelligence. At Apple, he was elevated to the role of Senior Vice President of Machine Learning and AI Strategy and reported directly to CEO Tim Cook.

Over nearly seven years, Giannandrea played several key roles:

• He unified scattered machine learning projects into a clearer structure under a central AI organization.
  
• He oversaw the development of Core ML (Apple’s machine learning framework), and guided AI work in areas such as image processing, recommendations, and on-device personalization.
  
• He pushed a strong privacy-focused vision of AI, emphasizing on-device processing to keep user data out of large centralized data centers.
  

According to Apple’s announcement and subsequent coverage, Giannandrea will remain at the company as an advisor until his retirement in the spring of 2026. That transition period is designed to maintain continuity while Subramanya settles into the role and ramps up new initiatives.

Tim Cook has publicly thanked Giannandrea for his contributions, particularly his work in building Apple’s internal AI capabilities and establishing the foundation upon which Apple Intelligence is being developed. The company is presenting the change as a natural handover at a time when its AI ambitions are expanding.

Siri, Delays, and the Pressure of the AI Race

Despite Giannandrea’s achievements, the pressure on Apple’s AI strategy has intensified over the last two years. Rivals have moved quickly conversational assistants based on large language models, generative search, and integrated AI “copilots” are now central selling points for competing platforms.

One of the most visible pain points has been Siri.

• Apple previewed a more capable, context-aware, and personalized version of Siri, but the rollout has been delayed multiple times.
  
• Reports have indicated internal concerns about whether the new Siri could meet Apple’s quality, reliability, and privacy standards on schedule.
  
• Earlier this year, some responsibilities for Siri’s day-to-day development were shifted away from Giannandrea, adding to speculation about a broader leadership change.

At the same time, Apple has been trying to integrate generative AI in a way that fits its identity:

• Many of its Apple Intelligence features are designed to run partly on-device and partly in secure data centers, with strong controls around data usage.
  
• The company has faced a delicate balance between moving fast enough to remain competitive and avoiding rushed releases that could damage trust.

In that context, the decision to bring in Subramanya is not just about replacing one executive with another. It is a response to a broader strategic question: how can Apple match or surpass competitors in generative AI while staying true to its privacy-first positioning?

What Changes for Apple’s AI Strategy Now?

With Subramanya in place and Federighi overseeing the broader software integration, Apple’s AI structure is becoming more streamlined. Several trends are likely to shape the next phase:

1. More tightly integrated AI in core products
   AI capabilities are expected to be less of a separate “feature” and more of a built-in layer across iOS, macOS, and other platforms. That includes smarter text tools, better content understanding, and context-aware assistance inside native apps and services.
   
2. A renewed focus on Siri
   The company has already confirmed that a more personalized Siri experience is targeted for release over the coming year, with deeper upgrades extending into 2026. Subramanya’s success will be judged heavily on whether Siri finally feels as flexible, conversational, and reliable as rival assistants—while still operating within Apple’s strict privacy boundaries.
3. Balancing on-device and cloud AI
   Apple’s chips are now powerful enough to handle substantial AI workloads on device. Even so, some generative AI tasks require data center-scale resources. Under Subramanya’s leadership, Apple will need to refine which tasks stay on the device, which move to the cloud, and how to make that split invisible to users while maintaining strong privacy guarantees.
   
4. Competing in a hot talent market
   The AI field is intensely competitive. Other tech giants have been hiring away experienced researchers and engineers from Apple, while Apple itself is recruiting specialists from rivals. The hiring of a senior leader from Microsoft who previously spent years at Google shows that Apple intends to stay aggressive in securing top AI talent.

A Strategic Turning Point for Apple

This leadership change comes at a turning point for Apple. AI is no longer just one feature among many; it is increasingly the layer that ties together search, apps, devices, and services.

Amar Subramanya arrives with direct experience running engineering for a major AI assistant at Google and helping to steer AI strategy at Microsoft. John Giannandrea leaves behind a more mature AI organization, a privacy-centric philosophy, and the early foundations of Apple Intelligence.

The next few years will show whether this combination—new leadership on top of a strong base—can deliver what Apple needs:

• A Siri that finally feels modern and genuinely helpful.
  
• AI features that are competitive with the best in the industry, but still aligned with Apple’s emphasis on user control and privacy.
  
• A cohesive AI story across iPhone, iPad, Mac, and new product categories.
  

For now, Apple is signaling that it is prepared to rethink leadership, structure, and strategy to make AI central to its future—and to ensure it is no longer perceived as lagging behind in one of the most important technology shifts of this decade.