Apple and Google have confirmed a multi-year partnership that will see Google’s Gemini models underpin a range of future Apple Intelligence features, including the next-generation version of Siri. The announcement marks a notable shift in how Apple plans to scale its artificial intelligence capabilities, while still keeping its core emphasis on on-device processing and privacy protections.
Under the agreement, Apple’s upcoming Foundation Models will be built using Google’s Gemini large language models and cloud infrastructure. These models will serve as the backbone for new Apple Intelligence features expected to roll out later this year, most prominently a more personalized and context-aware Siri. Apple said the decision followed internal evaluations that concluded Google’s AI stack offered the most capable base for its next phase of development.
Despite the collaboration, Apple emphasized that Apple Intelligence will continue to run primarily on Apple devices, supplemented by its Private Cloud Compute system. According to the company, user data will remain protected under the same privacy standards it currently applies across its ecosystem. In practical terms, Gemini will function as the underlying intelligence layer, while Apple retains control over how features are deployed, processed, and presented to users.
The move is expected to significantly enhance Siri’s capabilities. Gemini’s language models are considerably larger and more mature than Apple’s in-house equivalents, particularly in areas such as natural language understanding, contextual reasoning, and multi-step task handling. As a result, the upcoming Siri update is likely to feel closer to modern conversational assistants, narrowing a long-standing gap between Apple’s voice assistant and rivals.
Apple is widely expected to debut the updated Siri alongside the release of iOS 26.4, which is currently anticipated to reach the public in the March or April timeframe. The company first previewed its vision for a more personalized Siri and broader Apple Intelligence features at WWDC 2024, but development delays pushed back the rollout. Today’s announcement effectively confirms that those delays led Apple to seek external support rather than rely solely on internal model development.
The partnership also raises questions about Apple’s existing relationship with OpenAI. Apple currently integrates ChatGPT into Siri and Apple Intelligence for certain queries. While Apple has stated that there are no immediate changes to that agreement, it remains unclear how the Gemini integration will coexist with ChatGPT over the longer term or whether Apple will eventually streamline its AI partnerships.
Reaction from the wider industry has been mixed. Some observers see the deal as a pragmatic move that prioritizes user experience over platform rivalry. Others, including competitors, have raised concerns about increased concentration of AI power. Elon Musk, whose company xAI develops the Gemini rival Grok, publicly criticized the partnership, arguing that it strengthens Google’s influence across multiple major platforms.
Ultimately, the collaboration highlights the growing difficulty of developing competitive large language models independently. For Apple, leaning on Gemini appears to be a calculated decision to accelerate its AI roadmap while preserving its product philosophy. Whether users notice a clear leap forward when the new Siri arrives will likely determine how successful that strategy proves to be.
