Launch of the first processor based on ARM DynamIQ technology
Providing new IP such as computing libraries and comprehensive developer tools

Next-generation CPU and GPU IPs designed to accelerate the artificial intelligence (AI) experience have been unveiled.

ARM launched the Cortex-A75 and Cortex-A55, the first processors designed based on ARM DynamIQ technology announced last March, and also introduced the ARM Mali-G72 based on the Bifrost architecture.

ARM DynamIQ technology, first introduced in March, was the first milestone toward distributed intelligence from chip to cloud, and ARM Cortex-A75 and Cortex-A55 processors designed based on it have set a goal to improve AI performance by 50x within the next 3-5 years and are equipped with dedicated instruction functions required for AI performance tasks through DynamIQ technology.

It improves multicore capabilities and flexibility with DynamIQ big.LITTLE in a single computing cluster, and firmly defends end-device SoCs with ARM TrustZone technology, the security foundation for billions of devices. Functional safety has also been improved for Advanced Driver Assistance Systems (ADAS) and autonomous driving.

ARM Mali-G72, a new premium product from ARM Mali, the industry leader in shipments, further optimizes the SoC for distributed intelligence and device-based machine learning (ML). Mali-G72 is designed for high-fidelity mobile gaming, mobile virtual reality (VR), and new use cases for machine learning.

According to the results of a recent survey, 85% of consumers worldwide are concerned about the security of AI. This indicates the need for important indicators that devices must process and store more personal data to instill trust in AI's privacy protection.

Nandan Nayampally, General Manager of ARM Compute Products Group, stated, “Given that ARM technology has now reached 70% of the global population, realizing secure ubiquitous AI is a fundamental guiding design principle for ARM. As such, ARM has a responsibility to fundamentally reshape the computing experience for AI and other human-like computing experiences. To this end, ARM must realize faster, more efficient, and secure distributed intelligence between network-end computing and the cloud.”