Article Complete
AIM
When You Say ‘Hey Siri’, Google's Gemini May Respond Back
When You Say ‘Hey Siri’, Google's Gemini May Respond Back
Siri's New BFF

When You Say ‘Hey Siri’, Google's Gemini May Respond Back

Bloomberg reported that Apple plans to pay Google nearly $1 billion per year to integrate Gemini AI models into Siri.

Published Updated

Key takeaways

  • Apple plans to pay Google $1 billion annually to integrate Gemini AI into Siri.
  • Gemini AI aims to resolve ongoing issues with Siri's performance and delays.
  • Apple has delayed the upgraded Siri release to 2026 amid internal dissatisfaction.
  • Leadership in Siri development has shifted to enhance oversight and improve functionality.
  • Apple is developing its own trillion-parameter AI model for future consumer applications.
5 min

Apple is planning to pay Google $1 billion a year to integrate the Gemini AI model into Siri, according to a report by Bloomberg on November 5. 

This aims to address persistent issues with Apple Intelligence and the significant delays in the anticipated Siri upgrade and overhaul.

Citing sources familiar with the matter, the report stated that the companies are now finalising an agreement after an extensive evaluation period. Gemini will be deployed on Apple’s privacy-focused Private Cloud Compute servers.

In 2024, Apple announced an upgraded version of its voice assistant Siri, which uses AI to provide personalised responses, visual context integration, in-app actions, and more. While it was initially announced for this year, Apple has further delayed its release to 2026. 

Several engineers from Apple reportedly expressed dissatisfaction with the performance of the upgraded version of Siri that was being developed. 

Besides, Apple also internally re-shuffled its leadership in AI, moving the responsibility for Siri development away from AI chief John Giannandrea and placing it under the oversight of software head Craig Federighi, with Mike Rockwell (formerly of the Vision Pro team) becoming directly responsible for the assistant’s turnaround.

Currently, Apple Intelligence, the suite of AI features on Apple devices, is powered by in-house models — while the company is also building a ‘trillion’ parameter model in an attempt to catch up with successful large language models in the competition.

This is set to be available for consumer applications ‘as early as next year’ reported Bloomberg. 

“The hope is to use the [Gemini] technology as an interim solution until Apple’s own models are powerful enough,” added the Bloomberg report. 

Additionally, Apple is already in a partnership with OpenAI to integrate ChatGPT as an extension for Siri. 

If its partnership with Google materialises, it will be interesting to see how both these integrations operate and evolve. 

On the other hand, Google is paying Apple billions of dollars a year — $20 billion in 2022, for instance, to ensure that its search engine appears as the default on Apple’s Safari browser. 

However, Bloomberg stated that this partnership is going to be different, as the Gemini integration works on the back end without revealing any of Google’s branding to the users.