Computer chip that mimics human brain could power everyday devices

A Canterbury physicist has created a computer chip that mimics the brain and could change the face of the industry.

A Canterbury University scientist has developed a neuromorphic computer chip. Source: 1 NEWS

The prototype uses technology that can’t be seen by the human eye, but within a decade could be powering everyday devices.

"It's the only microscope in New Zealand that can see individual atoms," says Canterbury University physics Professor Simon Brown.

Costing three quarters of a million dollars, the microscope is an important piece of equipment when working with nanotechnology.

"You can't see anything with the naked eye or even with an ordinary microscope," says Professor Brown.

Physics Professor Brown is developing a neuromorphic computer chip.

"The nano parts we are using to build this chip are a hundred thousand times smaller than a hair and we are just spray painting them, scattering them on to a surface," he says.

It works like the human brain, and Professor Brown says it's smaller, faster and uses less power than anything else on the market.

"It's not going to learn to control us and take over the world - but it's capable of thinking in the sense of recognising a pattern."

It’s likely to be around a decade before it's fully functional, and its future could be in your phone.

“It's difficult to predict at the moment if this chip will be better at reading a fingerprint or a retinal scan when you sign into your phone.”

The prototype chip is this tiny dot right in the centre and it works very much like the brain, mimicking what neurons do to process and transmit information.

If successful it could be a global game changer.

"We've put in place a number of strategies around IT protection globally because we anticipate this is unique," says Business Development manager David Humm.

NZTech CEO Graeme Muller says this technology could be massive.

"If we could patent it in New Zealand it could be a billion dollar industry for us."

The next step is to make the chip recognise patterns, pictures or sound.

A prototype should be ready next year.