Google is serious about innovation. If the introduction of self-driving cars and balloon-powered Wi-Fi is not innovative enough for some, the search company really pushes its limits by trying to build its own quantum processors.
Last year, Google spent a total of $8 billion on research and development. This year, it is concentrating more resources on its Quantum Artificial Intelligence group, which is tasked with the mission of creating new processors and hardware that can conduct extremely fast calculations based on the principles of quantum physics.
Unlike conventional computers, which use transistors to represent the 1s and 0s in standard binary computing, quantum computers use quantum bits or qubits that can theoretically exist in multiple states. While a 1 will always be a 1 and a 0 will always be a 0 in conventional computing, in quantum computing, a 1 can be a 1, a 0 or both, allowing for much faster calculations and better processing power.
However, qubits are highly unreliable, and even the littlest changes in temperature or magnetism can alter their state altogether, which is why Google is bringing in more experts already experienced in establishing reliability. The experts come from a group of quantum researchers from the University of California, Santa Barbara led by physicist John Martinis and known for breaking new grounds in quantum computing. Specifically, Martinis and his team developed a new array of qubits that they say is more reliable and could lead the way to the creation of superfast quantum computers.
Google has already partnered with D-Wave Systems, a British Columbia-based quantum computing company which sold Google and NASA the Vesuvius machine in 2013, which Google says will be upgraded to a Washington processor with 1,000 qubits. Although some scientists have questioned the quantum nature of D-Wave's computers, noting that the systems are no faster than a traditional computer at times, D-Wave has maintained that it has built the first commercially viable quantum systems. Google says it will continue to work with D-Wave scientists along with the new team of researchers.
"With an integrated hardware group the Quantum AI team will now be able to implement and test new designs for quantum optimization and inference processors based on recent theoretical insights as well as our learnings from the D-Wave quantum annealing architecture," says Hartmut Neven, director of engineering at Google in a blog post.
It's no surprise why Google wants to build its own quantum processors, as it has been pushing its way to the forefront of ground-breaking AI technologies, including its driverless cars, robots and smartphones. However, the restrictions of conventional computing have prevented Google from coming up with smarter solutions to AI problems blocking its new technologies from going mainstream. With superfast quantum computing capabilities, Google's innovations could actually become everyday household instead of futuristic fantasies.
