Google's 53-qubit Sycamore and IBM's 50-qubit prototype: how the quantum milestone race unfolded

Google called it quantum supremacy. IBM called it a laboratory experiment with no practical applications. Both were describing the same 2019 result, and both had spent the preceding years building toward it.

According to a paper Fortune obtained in September 2019, Google's Sycamore processor contained 53 qubits. The company said the machine took about 200 seconds to sample one instance of a quantum circuit a million times. A state-of-the-art supercomputer would need approximately 10,000 years for the equivalent task. The researchers estimated the same job on a Google Cloud server would run for 50 trillion hours. On the quantum processor, they said, it took 30 seconds.

The paper appeared on NASA.gov before being taken down. Fortune reported that the Financial Times first broke the news. A Google spokesperson declined to confirm the authenticity of the paper and its results, and NASA did not respond immediately to a request for comment. A source at Google familiar with the situation suggested NASA accidentally published the paper early, before the team's claims could be vetted through scientific peer review, a process that could take weeks or months.

From six qubits to a supremacy target

The result did not arrive out of nowhere. In April 2017, MIT Technology Review reported that John Martinis, who leads Google's quantum research group, had given himself until the end of that year to demonstrate quantum supremacy. He defined it as a calculation beyond the reach of any conventional computer. Martinis said the experiment would pit Google's chip against one of the world's largest supercomputers.

At that point Google's newest chip had six qubits, arranged two by three. Martinis said the layout showed the technology still worked when qubits sat side by side, as they would in larger devices. The chip also tested a manufacturing method in which qubits and their conventional control wiring are made on separate chips and later bump bonded together. The approach was meant to remove extra control lines that can interfere with qubit operation.

"That process is all working," Martinis told MIT Technology Review. "Now we're ready to kind of move fast." He said designs for devices with 30 to 50 qubits were already in progress, and that he would need a grid of 49 qubits for the supremacy experiment. Google had previously released results from a nine-qubit chip arranged in a line.

Martinis joined Google in late 2014 from the University of California, Santa Barbara, where he remained a professor. By the time of the 2017 article, his group numbered roughly 25 people. MIT's Simon Gustavsson told the publication that Google was one of the leaders in the field and that the race was "pretty comparable between Google and IBM."

Chris Monroe, a University of Maryland professor and cofounder of the startup IonQ, was more measured. "It'll be an academic milestone," he said of the supremacy goal. "Afterward you still have to figure out how to make it more scalable and programmable." Martinis agreed much remained to be done. He argued the experiment could become a benchmark for anyone claiming a working quantum computer.

IBM's 50 qubits, and a dispute over the word supremacy

IBM moved first on the number that Google had targeted. In November 2017, the Associated Press reported that IBM scientists had built and measured a processor prototype with 50 qubits. Dario Gil, who led IBM's quantum computing and artificial intelligence research division, said it was the first time any company had built a quantum computer at that scale.

Seth Lloyd, an MIT mechanical engineering professor not involved in IBM's research, told the AP that IBM likely still had glitches to work out, but that the 50-qubit announcement was a sign of significant progress.

Less than a month later, at the Q2B conference at NASA Ames, Martinis pushed back on qubit counting as a scoreboard. "Press releases always talk about quantum space race in number of qubits," he said, according to Gizmodo. "It's more than just quantity, it's qubit quality." Google, he said, was fabricating its 49 or 50 qubit supremacy device that month and would begin testing within two weeks, Christmas vacation delays aside.

Gizmodo's report described the problem Google intended to run: entangle the qubits, let the system evolve, then measure. The final state takes different values with different probabilities. Working out the possible outcomes is so complex that a classical computer has to simulate the quantum machine, potentially taking weeks to do what the quantum computer does in minutes. The exercise tests not only qubit count but qubit quality, since the qubits must not produce wrong values or decay into ordinary bits by interacting with their environment.

The conference drew representatives from automotive and airline companies, banks, software firms and the military. Volkswagen, Airbus, Citibank, Emerson and the European IT company Atos were among those present, alongside venture capital firms. John Preskill, the CalTech theoretical physicist, told the audience the field was entering what he called the NISQ era, for noisy intermediate-scale quantum computers. He said progress toward a fault-tolerant machine must continue.

Quantum Volume and the decade-long wait

IBM's answer to qubit counting arrived in March 2019. At the American Physical Society March Meeting, the company outlined its highest Quantum Volume to date, a metric that combines qubit count, connectivity and coherence time while accounting for gate and measurement errors, device cross talk and compiler efficiency.

IBM said its Q System One, built around a 20-qubit processor, produced a Quantum Volume of 16, double the then-current IBM Q figure of 8. The company also said the system had some of the lowest error rates it had measured.

ZDNet reported IBM's projection that Quantum Volume would need to double every year to reach Quantum Advantage within the next decade. Quantum Advantage is the point where quantum applications deliver significant advantages over classical computers. IBM has doubled the power of its quantum computers annually since 2017, the company said. It made its quantum technology available through a cloud service in 2016 and has worked with partners on business and science use cases.

That framing set up the argument that followed Google's September 2019 paper. Dario Gil, by then head of IBM Research, advised against using quantum supremacy as a progress metric. "The experiment and the 'supremacy' term will be misunderstood by nearly all," he told Fortune, describing the work as a highly special laboratory experiment with no practical applications. "Quantum computers will never reign 'supreme' over classical computers, but will rather work in concert with them, since each have their unique strengths."

Jim Clarke, Intel Labs' director of quantum hardware, called Google's update "a notable mile marker" and said a commercially viable quantum computer would require many more research and development advances. "While development is still at mile one of this marathon, we strongly believe in the potential of this technology," he said.

What the paper claimed, and what it did not

The Google researchers wrote that quantum speedup was achievable in a real-world system and was not precluded by any hidden physical laws. They predicted quantum computing power would grow at a double exponential rate, faster than the exponential rate that defined Moore's Law. They also wrote that quantum computing was transitioning from a research topic to a technology that unlocks new computational capabilities, and that "we are only one creative algorithm away from valuable near-term applications."

The machine itself was a step down in qubit count from an earlier design. Google had previously designed a 72-qubit device named Bristlecone, but scaled back to the 53-qubit Sycamore for the experiment.

The company's earlier public statements are worth holding next to the 2019 claim. In 2017, Martinis said quantum processors would need to be much larger than 50 qubits to be capable of useful work. Monroe's assessment that supremacy would be an academic milestone, with scalability and programmability still to be solved, was not contradicted by anything in the paper.

Google's own spokesperson declined to confirm the paper's authenticity, and NASA, which had briefly hosted it, stayed silent. Peer review, the company source said, could take weeks or months. The result that reached the public first arrived through a leak, not a journal.

That is the shape of the milestone race as the record shows it: a 50-qubit prototype from IBM in November 2017, a 53-qubit supremacy claim from Google in September 2019, and a metric dispute in between about whether qubit count or qubit quality is the number that matters. Businesses, as Gizmodo noted from the Q2B floor, were told practical use could be decades away. IBM's own projection put Quantum Advantage within a decade if Quantum Volume kept doubling every year.

Both companies kept building. The argument about what to call the result will likely outlast the hardware that produced it.