In an official press release issued earlier today, Apple revealed that John Giannandrea, a prominent figure in the field of applied machine learning and artificial intelligence, will be retiring from his position as the company’s Senior Vice President responsible for machine learning and AI strategy. Giannandrea, who joined Apple in 2018 after a distinguished tenure at Google, was widely regarded as one of the most respected minds in the AI sphere. His departure signals the end of a significant chapter for Apple’s AI division, one that coincided with a period of both ambitious experimentation and public scrutiny. Notably, this change comes at a time when Siri—Apple’s most recognizable and once trailblazing digital assistant—has come to symbolize the company’s lagging reputation in the rapidly evolving AI landscape, appearing to many as a vestige of the pre-ChatGPT era rather than a leader in it.
Alongside this announcement, Apple also disclosed that it has appointed Amar Subramanya as its new Vice President of Artificial Intelligence. His arrival, reportedly following his tenure as Corporate Vice President of AI at Microsoft, marks a deliberate effort by Apple to revitalize its AI strategy with fresh leadership. The statement emphasized that Subramanya’s role will be central to Apple’s ongoing innovation initiatives and will play a pivotal part in shaping the company’s forthcoming “Apple Intelligence” features—a suite designed to integrate generative AI more deeply across its devices and services.
The broader message of Apple’s release extends beyond Siri alone, a product that predates Giannandrea’s tenure by nearly a decade. Despite Apple’s recent public demonstrations of generative AI capabilities through its Apple Intelligence platform, public perception continues to conflate Apple’s AI identity almost entirely with Siri’s performance and limitations. During Giannandrea’s leadership, the virtual assistant became emblematic of what many critics viewed as Apple’s cautious and somewhat outdated approach to AI innovation.
Still, Siri has had supporters who argue that its criticism is often overstated. Just days ago, YouTube technology commentator Marques Brownlee published a video titled “Siri Isn’t That Bad,” cautiously defending the assistant’s usefulness within its designed constraints. He echoed a subset of users who maintain that Siri performs adequately when engaged appropriately—that is, when users frame their requests within the system’s intended command structure rather than expecting open-ended conversational intelligence.
Nevertheless, comparisons to modern large language models (LLMs) have underscored how far Siri trails behind newer AI systems that handle complex, context-aware reasoning. In fact, Apple has already integrated external generative AI support into Siri: beginning with iOS 18.2 approximately a year ago, Siri has been quietly handing certain queries—such as creative storytelling or extended conversation prompts—directly to ChatGPT. For instance, if a user asks Siri to craft a bedtime story about a duck for a five-year-old child, the request may be rerouted, allowing ChatGPT to generate the response on Siri’s behalf. While this partnership provides functionality users increasingly expect, it also implicitly highlights the gap between Apple’s proprietary models and the competition’s more advanced tools. Such developments undoubtedly carry personal resonance for Giannandrea, who was originally hired to close precisely this competitive gap.
As journalist Tom McCay wrote in 2018, Apple’s recruitment of Giannandrea—then Google’s Chief of Search—was seen as a bold attempt to accelerate its AI capabilities to the level of Google Assistant or Amazon’s Alexa, both recognized at the time for their sophistication and adaptability. Yet, seven years later, Siri finds itself relying on an external system as a fallback mechanism, invoking ChatGPT when Apple’s in-house solutions fall short. This reality starkly contrasts with the vision that fueled Giannandrea’s hiring.
Recent months have brought further signs of strategic transition. Earlier in the year, reports surfaced indicating that Giannandrea was no longer overseeing the Siri team directly. Then, last month, leaks suggested that Apple is planning to license a foundational AI model from Google to serve as the core technology underpinning the next-generation version of Siri—a partnership that illustrates Apple’s willingness to collaborate externally to remain competitive. While Subramanya’s title differs slightly from Giannandrea’s, his extensive experience—particularly his involvement in Google’s Gemini project during his sixteen-year tenure at the company—positions him as a pivotal figure in defining Siri’s future architecture. These credentials suggest that Apple’s upcoming digital assistant will likely carry traces of his technical philosophy and engineering influence.
Despite these strategic adjustments, many observers continue to perceive Apple as following rather than leading the AI race. Siri was initially conceived in an era when simply executing deterministic responses to a narrow range of commands represented a marvel of consumer technology. Over time, it has expanded modestly, gaining additional functions but rarely exceeding expectations established by rival platforms. Under Giannandrea’s leadership, Apple’s virtual assistant remained serviceable yet consistently overshadowed by the more dynamic, conversation-driven assistants powered by large-scale language models. With Subramanya now assuming control, it appears that Siri may soon evolve into a system capable of handling open-ended LLM-powered interactions—essentially the very class of tasks Apple had long deferred.
Should this transformation enable Siri to finally meet the capabilities expected of voice assistants in, say, 2026, it will undeniably represent progress. Yet the fundamental question remains: how long will merely catching up suffice in a field advancing at exponential speed? Apple’s commitment to innovation is unquestioned, but its ability to redefine user expectations—as it once did repeatedly—will determine whether this next version of Siri feels like a breakthrough or simply a belated correction.
According to Apple’s official statement, John Giannandrea’s retirement will formally take effect in the spring of next year, closing out a significant chapter in Apple’s ongoing pursuit of artificial intelligence excellence.
Sourse: https://gizmodo.com/apple-has-announced-the-retirement-and-apparent-replacement-of-the-guy-in-charge-of-siri-2000693767
