Meta is not a fan of quantum computing–at least not right now. That much could be gleaned from comments made by Meta’s Chief Scientist Yann LeCun (and reported by CNBC) at a company event in San Francisco earlier this month. At that event, LeCun cast doubt on the “practical relevance” of quantum computing.
In this, Meta, parent of Facebook, might be virtually alone among high-tech giants, as numerous recent high-profile quantum computing announcements from a host of big-name firms–Amazon Web Services, Microsoft, Google, Nvidia, IBM, and Intel–suggest that quantum computing is knocking at the door of practical and commercial relevance.
Among recent notable quantum moves:
-
IBM this month unveiled its Quantum System Two quantum computer, powered by three of its 133-qubit Heron processors, a milestone that IBM said ushers in “the era of quantum utility,” with Heron featuring “five-fold error reduction” over previous generation quantum processors. Katie Pizzolato, IBM’s director of Quantum Theory and Computational Science, said during that announcement, “Quantum utility means we’ve reached that point at which a quantum computer is able to perform reliable computations at a scale beyond brute force classical computing methods.” IBM also unveiled a roadmap that takes its quantum processor from error reduction efforts to full “error correction” by 2029.
-
Just days before IBM’s announcement, AWS unveiled what it described as a quantum error correction chip. The research prototype was developed and manufactured at the AWS Center for Quantum Computing on the campus of the California Institute of Technology (Caltech).
-
Microsoft last month announced a partnership with quantum computing start-up Photonic to work on quantum networking, a move that came several months after Microsoft reported a new way of building logical qubits–error-corrected qubits derived from physical qubits.
-
Google this month discussed its research on a new quantum algorithm, which came after several announcements during 2023 about the company’s efforts in error suppression and correction.
-
Nvidia has been mining the intersection of quantum and classical computing, where quantum simulation software often is being run on classical machines to suggest the game-changing capability that quantum computers eventually will have. After taking a major step toward GPU-accelerated quantum computing earlier this year, Nvidia last month highlighted partnerships and projects that are advancing the notion of classical-quantum hybrid computing.
-
Intel released its Tunnel Falls 12-qubit chip to the research community in mid-2023. While Intel has made no further comments yet about broader market availability, this move represented a major step forward for a company that previously had been quiet about its quantum efforts.
Who is missing from the list above, other than Meta? AMD has been fairly quiet about its quantum activities, but has acknowledged work with quantum computing companies like Riverlane. Also, Broadcom has been similarly quiet, but recently engaged in a partnership with Caltech to establish a quantum lab, a collaboration “supported with a significant investment from Broadcom,” according to a Caltech press release (Broadcom did not put out its own press release about the partnership.)
Though AMD and Broadcom have not promoted quantum computing as vigorously as IBM, AWS, Nvidia, and others, they also have not downplayed it in the way that Meta chose to do.
Many of the moves listed above demonstrate work being done to address quantum computing’s biggest current weakness and the thing most keeping it from broader commercial relevance–its error-prone nature (although “error-prone” in this context often represents reliability a couple of percentage points below 99%+).
But, in fact, 2023 also has been a year of previously unforeseen commercial activity for quantum–as in the actual selling and buying of quantum processors and full-scale machines. For example, pure-play quantum computing start-up IonQ said earlier this year it had sold two full systems to a European lab, and predicted more sales activity into next year. Meanwhile, fellow pure-play quantum start-up Rigetti Computing said around the same time that it has sold quantum processors to two clients, and the firm this month formally unveiled a commercially available 9-qubit quantum processor, priced starting at $900,000 and available for shipping in four to six weeks.
All of this represents a lot of evidence that quantum computing is just around the corner from practical and commercial relevance–if not already there for some of the firms looking to make a living off of it. Meta may be wrong. Perhaps, in the wake of over-committing to the metaverse, it is just looking to avoid another too-early, too-energetic public over-commitment to a relatively new technology. But if Meta doesn’t believe in quantum computing right now, don’t be surprised if things change very quickly in 2024.