Apple has announced a significant deal with OpenAI to integrate its AI model into Siri, but the fine print reveals that Google has also played a crucial role. To build its AI models, Apple used its own software framework along with Google’s tensor processing units (TPUs) available via Google Cloud. Google’s TPUs, designed specifically for AI applications, offer performance competitive with Nvidia’s AI chips. Google, which has been developing TPUs for around a decade, plans to launch a sixth generation this year. While Apple didn’t specify the extent of its reliance on Google’s hardware compared to other vendors, using Google’s chips generally involves purchasing cloud access similar to services from Amazon AWS or Microsoft Azure. Both Apple and Google declined to comment further.
Source – CGTN