Just as graphics got offloaded to a dedicated processor – soon AI might have it’s own processing unit too.

Bloomberg reported that Apple is developing a processor specifically devoted to AI tasks. Reportedly known as the Apple Neural Engine, it’s noted to handle tasks such as  facial recognition and speech recognition.

Within AI usage and implementation – Apple is notably lagging behind its peers – or rivals – in Amazon and Alphabet. Despite the headstart that Siri had in voice recognition, Amazon’s Alexa and Google’s Assistant has visibly taken over with capabilities, leading the deployment of AI into their products.

While currently Apple, like many other manufacturers processes most things on a main processor and a separate graphics chip, a new chip dedicated for AI will allow for improved battery performance. This is not unlike Qualcomm Snapdragon chips with a dedicated AI module, or Google’s TPU (Tensor Processing Unit). TPU was first used in Google’s datacentres to power search results and image-recognition. Recently at I/O conference this year, Google announced a new version that will be available to clients of its cloud business. Nvidia also sells a similar chip to cloud customers.

According to Bloomberg sources, the Cupertino giant plans to integrate the chip into many devices including the iPhone and iPad, and has started testing prototypes of future iPhones. There isn’t confirmation of whether it will make it in time for the launch of the 10th anniversary handset rumoured to take place this fall.

Apple also plans to offer developer access to the chip so third-party apps can also offload artificial intelligence-related tasks, according to Bloomberg’s report.

With plans on the way for Apple’s WWDC conference in June, we could see the chip being discussed in the annual developers conference. It is also expected that iOS 11, its new operating system for iPhones and iPads, will be introduced.