Apple names new AI chief with Google and Microsoft expertise as Giannandrea steps down

Reviewed byNidhi Govil

38 Sources

Share

Apple replaces AI chief John Giannandrea with Amar Subramanya, a former Google Gemini engineering leader who recently worked at Microsoft. The move follows persistent struggles with Apple Intelligence and a delayed Siri overhaul, signaling a strategic shift as the company attempts to catch up in the AI race.

News article

Apple AI Leadership Undergoes Major Transformation

Apple announced Monday that John Giannandrea is stepping down as the company's AI chief, a position he has held since 2018

1

. The AI leadership change marks the most visible shakeup in Apple's artificial intelligence division since the company launched Apple Intelligence in 2024

5

. Giannandrea will remain at Apple as an advisor through spring 2026 before retiring, but his replacement takes effect immediately

4

.

Amar Subramanya, the new AI chief, brings extensive experience from both Google and Microsoft. He spent 16 years at Google, most recently leading engineering for the Gemini Assistant, before joining Microsoft as corporate vice president of AI just four months ago

1

3

. As Apple's vice president of AI, Subramanya will report directly to software engineering SVP Craig Federighi, taking responsibility for foundation models, machine learning research, and AI safety and evaluation

4

.

Siri Setbacks Drive Leadership Restructuring

The timing of this AI leadership change reflects mounting pressure following the troubled rollout of Apple Intelligence. Since its October 2024 launch, the company's AI suite has faced harsh criticism, with reviews ranging from underwhelming to alarmed

1

. The notification summary feature generated embarrassing false headlines in late 2024 and early 2025, including BBC complaints about fabricated news stories involving Luigi Mangione and darts player Luke Littler

1

.

The Siri overhaul became particularly problematic. A Bloomberg investigation revealed that when Craig Federighi tested the new Siri weeks before its planned April launch, many touted features simply didn't work

1

. The Apple Intelligence delay triggered class-action lawsuits from iPhone 16 buyers who had been promised an AI-powered assistant

1

3

. Apple admitted in March that developing advanced Siri features was taking longer than expected, pushing the rollout to 2026

2

3

.

Organizational Issues Behind the Scenes

Before John Giannandrea stepping down was officially announced, Bloomberg reported that Tim Cook had already lost confidence in his AI leadership. In March, Cook stripped Siri oversight from Giannandrea entirely, handing it to Vision Pro creator Mike Rockwell

1

2

. Apple's secretive robotics division was also removed from his control

1

.

The investigation painted a troubling picture of organizational issues, including weak communication between AI and marketing teams, budget misalignments, and severe dysfunction

1

. Some employees mockingly referred to Giannandrea's group as "AI/MLess," while AI researchers departed for competitors including OpenAI, Google, and Meta

1

.

Google Gemini Integration and Strategic Pivot

In a striking development, Apple is reportedly leaning on Google Gemini integration to power the next version of Siri

1

2

. This represents a significant shift for a company locked in intense rivalry with Google across mobile operating systems, app stores, browsers, and smart home devices for over 15 years

1

.

Analyst Marina Koytcheva of STL Partners suggested the move signals Apple knows it has lost ground in the AI race, hiring Subramanya from AI leaders Microsoft and Google to catch up in AI race

4

. She warned that even Apple's substantial customer loyalty has limits, noting that users will demand the right experience or look elsewhere

4

.

Privacy-First Approach Faces Competitive Pressure

Apple has maintained a distinct strategy focused on on-device processing using custom Apple Silicon chips, avoiding data collection that competitors rely on

1

. When complex requests require cloud processing, Apple routes them through Private Cloud Compute servers that process data temporarily before immediate deletion

1

.

This privacy-first AI philosophy comes with clear trade-offs. On-device models remain smaller and less capable than massive models running in competitors' data centers

1

. Apple's reluctance to collect user data has forced researchers to train models on licensed and synthetic data rather than the extensive real-world information fueling rivals' systems

1

. Whether this approach will pay off or leave Apple permanently behind remains an open question as Subramanya and Federighi face the challenge of delivering improved AI capabilities while maintaining Apple's privacy commitments

1

4

.

Today's Top Stories

TheOutpost.ai

Your Daily Dose of Curated AI News

Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.

© 2025 Triveous Technologies Private Limited
Instagram logo
LinkedIn logo