Aryabhat-1 is a new AI-focused chip by Indian Institute of Science

A team at the Indian Institute of Science (IISc) has constructed the Aryabhat-1 (Analogue Reconfigurable Technology And Bias-scalable Hardware for AI Tasks) chipset prototype.

An IISc statement added that a chipset like Aryabhat-1 can be particularly useful for Artificial Intelligence (AI)-based applications such as object or speech recognition, or those that demand enormous parallel computing operations at high speeds.

Because the design process is straightforward and scalable, the majority of electronic devices, particularly those involving computers, employ digital chips.

The researchers claim that the benefit of analogue is enormous. Analogue computing has the potential to outperform digital computing in applications that do not require accurate computations, as analogue computing is more energy-efficient.

Also watch: AirPods Pro 2 tipped to come with a USB-C port

Despite this, there are several technological obstacles to overcome while creating analogue chips. Analogue processors are more difficult to test and co-design than digital circuits. By compiling a high-level code, it is simple to synthesise large-scale digital processors, and the same architecture may be transferred across successive generations of technological advancement.

Due to the inability of analogue chips to scale - they must be individually customised when migrating to the next generation of technology or a new application - their design is costly.

In order to overcome these obstacles, Aryabhat-1 uses a unique architecture that permits the building of analogue processors that scale similarly to digital processors. Their chipset is reconfigurable and programmable, allowing identical analogue modules to be migrated across several generations of process design and applications.

The researchers are planning to work with industry partners to commercialise this new technology.