That an all-powerful semi-conductor manufacturer would go deep on quantum computing seems foretold. This is, after all, the future, right? And, at least when it comes to continually shrinking chip processes, Intel is all about the future. With quantum computing, however, it isn't quite so easy.
On Friday, Intel CEO Brian Krzanich published an open letter pledging to invest $50 million in long-term quantum computing research. Keywords here are long-term: While 2022 seems like a reasonable outlook for 5 nanometer chip manufacturing nodes, quantum computers are at least 12 years away, according to Intel's estimates. The investment will put Intel into the same ring as IBM and Google, both of whom are pushing aggressively into fundamental quantum computing research.
Intel's investment will take the form of funding for QuTech, a research unit at the Technical University of Delft in the Netherlands, and will be doled out over the course of 10 years. "TU Delft has been working on the science behind quantum computing for many years and has great vision into the challenges around interconnects in particular," Krazanich writes.
How can we reliably fabricate, connect and control many more qubits?
The quantum engineering challenge is enormous, which he at least acknowledges: "How do we connect thousands of quantum bits, or qubits, together?. How can we control them? How can we reliably fabricate, connect and control many more qubits? Even measuring qubit signals is going to require an entirely new class of low temperature electronics that don't exist today."
Krazanich's honesty about QC and its myriad challenges is refreshing at a point of peak quantum hype, in which even the NSA is releasing statements bordering on paranoia. Thanks in part to D-Wave, the Vancouver, BC outfit made famous for pitching a quantum system that might not be a meaningful quantum system, the challenges and trials of producing a real quantum computer tend to get buried. In reality, qubits and quantum systems remain fundamentally precarious information-bearing entities, which greatly limits how we can put them together into a practical machine that can perform practical computations.
Worth noting, however, is that $50 million for Intel is pocket change. Consider the company's recently-constructed $5 billion plant in Arizona that is at the moment doing nothing more than collecting dust. For the big picture, it might be worth noting that in 2013 Intel took in some $56 billion in total revenue. By the time quantum technology has finally arrived, $50 million will seem like the best deal in the history of computing.