Timeframe aside, the prediction made sense for two reasons: 1). It was MIT's engineers and physicists that actually devised one of the first reliable methods for integrating this photonic circuitry onto a silicon chip and 2). Chip manufacturers are going to need to start figuring out alternate ways to squeeze out the needed performance from chips as electronic components (and the metal interconnects between them) continue to shrink in size.

Enter today's announcement on Intel's Research blog. The device Ansheng Liu is holding in the picture above is something called a laser modulator, a component that is capable of encoding optical data at 40 billion bits per second--a speed that conveniently matches the fastest devices deployed today using other materials, Liu said.

Mirroring MITs assertion, Liu said that chip makers have shown a keen interest in photonic integrated circuits (PICs) over the past few years because of their ability to provide a cost-effective solution for optical communication and future optical interconnects in the computing industry.

There are obstacles, though: One of the key components needed for silicon PICs is the very high-speed silicon optical modulator, which is used to encode data on optical beam. Unfortunately, most of today's commercially available optical modulators only achieve around 10 Gbps transmission speeds and tend to be based on "more exotic electro-optic materials" such as lithium niobate and III-V compound semiconductors, Liu said.

But with the successful demonstration of Intel's new silicon modulator along with the electrically pumped hybrid silicon laser, Liu says it will now be far easier to integrate multiple devices on a single chip that can transmit terabits of data per second in the near future.

In short, five years might just be about right.