SoftBank-owned Arm's latest chip designs feature improved AI performance on edge IoT devices

Chip designer Arm unveiled two new chip designs that can provide better AI functionality on edge Internet-of-Things (IoT) devices, which typically lack robust power sources.

The Cortex-M55 processor delivers machine learning performance that's up to 15 times better than its previous iterations, and a five-fold improvement in digital signal processing performance. And the Ethos-U55 NPU, the industry's first microNPU, is designed to pair with the Cortex-M55 to handle heavier workloads.

Together the two chips can provide a 480-fold uplift in machine learning performance compared with previous units while maintaining a small, power-efficient package. When the chips hit the market in 2021 they will enable tech companies to implement AI applications locally on smaller devices without sacrificing performance or size.

The new chips will be a boon for companies working on data-intensive on-device AI applications. On-device processing means that an AI application, like a voice assistant, can operate without a stable internet connection, thereby greatly improving its functionality.

Because data is stored and processed locally, on-device AI can provide quicker results, as it doesn't have to transmit data to and from the cloud to be processed - data gains roughly 0.82 milliseconds of latency for every 100 miles it travels. Fast processing speeds are critical in applications like voice assistants, where any lag in response time can harm the user experience.

Additionally, because data is stored locally it can be kept safe in the event of a cloud data breach, a major concern among IoT users. Over 90% of US Fortune 500 firms have experienced a cyber attack against their cloud systems, with 60% experiencing a data breach, according to a Proofpoint study cited by TechXplore.

Arm's newest chips are evidence of the impact that manufacturers' demand for diverse voice form factors is having on the connectivity ecosystem and supply chain. Tech giants are creating a range of voice-enabled products in alternative form factors, such as ceiling lights and wall clocks.

Companies are even creating battery-powered devices with voice assistants, which face power constraints. This push to add AI functionality to smaller and more power-constrained devices is pressuring component vendors to keep pace with the evolving needs of device makers.

Chip designers and manufacturers have a large incentive to produce the most efficient and compact chips, as they stand to cash in on a lucrative opportunity - in 2020 alone, chipmakers are expected to generate almost $2.6 billion by shipping 750 million edge AI chips, a number which is expected to double by 2024, per Deloitte.