In March, Patently Apple posted a report titled "Apple Chips will be advancing to the Armv9 Architecture, the first major update in a decade that will advance AI, AR/VR, 5G and Security." The report noted that "As far as AI processing goes, Scalable Vector Extension 2 (SVE2) would be a part of the Armv9. The technology has been designed so that digital signal processing and machine learning tasks can be carried out in a better manner. It is expected that AR, VR, 5G systems and ML workloads, including voice recognition and image processing, would all be benefitted by this."

Another key aspect of the new ARMv9 architecture is advancing security and "confidential computing." Apple is the leading company in Silicon Valley that is determined to protect user privacy and the new ARM architecture will provide deep confidential computing attributes.

In a new Arm press release late yesterday, they introduced their "Arm Confidential Compute Architecture (Arm CCA)," a key feature of Armv9-A and the next step in transforming how we think about the trust model of compute environments in every application.

Yesterday they unveiled the initial technical specifications for Arm CCA and sharing their vision for the next steps to unlocking the power and full potential of data, by making the benefits of confidential computing accessible to all developers.

Currently, applications and virtual machines place huge amounts of trust in the supervisor software (kernels or hypervisors) that manage them. Supervisors can access the resources used by applications for their program code and data. Exploits against supervisors can therefore leak confidential data or algorithms held in the applications.

Confidential computing changes the traditional trust relationship between applications and supervisors by removing the supervisor’s right to access the resources used by the application, while retaining the right to manage them. Removing that right of access is critical because the devices we use today handle large quantities of confidential data. Cloud systems can be running payloads from many different customers, while mobile devices can contain both personal and business information, from medical data to company emails. Confidential computing reduces the need to trust unseen technology within any compute environment.

Ron Martino, Executive Vice President and General Manager of Edge Processing, NXP Semiconductors: "In our increasingly automated world where data privacy and confidentiality are imperative, Arm CCA provides an opportunity to add a complementary layer of security to the integrated security capabilities that have been foundational to our processing portfolio for IoT, industrial, and automotive markets. Data generated, processed and sent by intelligent edge devices must be protected, and this is a step toward achieving that desired integrity and confidentiality.”

For more details on this, read the full Press Release.