It has been one year since Peter Chapman left e-commerce giant Amazon to become CEO of IonQ, a small five-year-old quantum computing startup. Chapman is a seasoned technology executive who brings a wealth of large system software, AI and computing experience to IonQ. At Amazon, he was one of two executives responsible for the complex, ever-changing mixture of technology and logistics that supported Amazon’s premiere service differentiator, Amazon Prime.
During Chapman’s year of leadership, IonQ has maintained a low profile. Still, judging from research papers, articles, and presentations during that period, the company has been busy. I also got the sense from a previous discussion that Peter Chapman doesn’t like to waste time broadcasting what he’s going to do. Instead, he does it without a lot of fanfare.
As its name implies, IonQ uses ions for quantum computing. An ion is made by removing an electron from an atom. Trapped ion technology was developed back in the 1990s and used for atomic clocks. Although many labs around the world are testing small trapped ion computers, Honeywell and IonQ are the two most prominent and successful commercial users of the technology.
Honeywell recently revealed its plans to implement a new trapped ion architecture in the coming months and years. In late 2019, Chapman successfully raised $55 million in funding for IonQ from Samsung Catalyst, Mubadala Capital and several other investors. I was curious and asked him how IonQ planned to use the capital over the next three to five years.
The answer surprised me. He told me that the funding enabled IonQ to start working on systems in parallel—three generations of hardware at the same time, to be specific. Creating the next generation of any quantum computer architecture is research-intensive and difficult-developing three generations at the same time is very difficult, very research-intensive and extremely ambitious.
This will be IonQ’s fifth, sixth and seventh generation of quantum computers. According to Chapman, the company’s prior generations followed the same development pattern. How is this strategy possible, and why does IonQ do things this way? Chapman explained, “None of our quantum computers have anything to do with any of the previous generations we’ve built. We’re building a system that’s made from off-the-shelf components made all over the world. I don’t think there’s one part identical to a previous generation.”
IonQ already put its first-generation trapped ion quantum computer out to pasture. They say dogs age seven years for every human year—the same rule seems to apply to quantum technology. Although IonQ’s first-generation quantum computer is only four-years-old, it is now an ornamental relic on display in IonQ’s main lobby.
IonQ’s fifth-generation computer is scheduled for testing and possible release sometime this summer. Chapman declined to reveal exact dates for its release or any details, such as the number of qubits. “We haven’t made that announcement, but we will shortly,” he said. “And it [the number of qubits] will be more than some of our competitors.”
It is IonQ’s practice to publish only qubit numbers that can practically be used. IonQ researchers experimentally loaded 160 qubits into a single ion trap, but that many ions proved too difficult to control to be useful. The IonQ scientists have also performed operations on as many as 79 qubits. In 2018, IonQ successfully used 11 qubits in its production trapped ion computer. I will remind all readers, I believe quantum volume, which I go into detail here, is a better indicator of useful performance than relying on raw qubits alone.
IonQ’s stated goal is to double the number of useful computational qubits every year. Long term, Chapman believes it will eventually be possible to have thousands of qubits on a single chip.
I asked Chapman what he thought about Google’s recent quantum supremacy achievement, which utilized 53 superconducting qubits. He stressed the company’s focus on the practicality of quantum computing over milestones like quantum supremacy.
That said, Chris Monroe, a founder of IonQ, believes IonQ researchers achieved quantum supremacy months before Google. It was the company’s choice not to get involved in a controversial intellectual debate.
IonQ expansion plans
One element of Chapman’s strategy is improvement of the existing infrastructure. To that end, IonQ’s quantum computers will be moved out of their current research environment into a more controlled and traditional data center environment equipped with upgraded security, power generators and battery backup.
The move is part of IonQ’s $6 million expansion of its headquarters. The plan also includes the construction of a new quantum lab and data center in partnership with the University of Maryland. Beefing up the infrastructure should improve IonQ’s ability to provide its customers with around-the-clock support and higher system availability.
A new data center also means more space for future research requirements. The center will be able to house at least sixteen quantum computers, and possibly more if the size of trapped ion quantum computers continues to shrink. The first trapped ion quantum computer was about the size of an average living room. Since then, according to Chapman, each new generation of trapped ion quantum computers has been significantly smaller than the previous generation. “We have them on an all-qubit diet,” he quipped.
In addition to pushing the computational power of IonQ’s quantum computers, Chapman also places a lot of emphasis on improving system uptime and performance. In December 2018, when IonQ’s first quantum processor came online, physicists had to manually hand-tweak and calibrate the system to achieve maximum performance. Today, all those processes are automated.
Customer service and cloud computing
Improving and maintaining a consistent level of customer service is also essential. Chapman plans to hire more staff for IonQ’s Seattle office in order to provide a greater focus on customer service.
The main reasoning behind the company’s presence in Seattle is its agreement to provide Amazon AWS and Microsoft Azure cloud access to its quantum computer. Amazon’s headquarters are in Seattle, while Microsoft’s headquarters are in Redmond, WA, only 15 miles away. In turn, cloud access allows IonQ to run thousands of daily jobs for customers participating in private beta tests.
Chapman believes that quantum computing on the cloud and the ability for people to get access to real quantum hardware is accelerating faster than anyone initially anticipated.
After spending an hour with him, I knew he was a visionary. He revealed that side of himself in one last quote I’d like to share: “There is one great thing I love about the quantum cloud environment, and that is the largest financial institution and a kid in a garage have equal access to it. It’s democratic. You don’t need a $200,000 quantum club card to get access to one of these things. And, if we’re looking for another Dan Bricklin, quantum cloud computing makes it just as likely to come from the kid in a garage as it does from a government lab or a large company.”
From an analyst’s perspective, IonQ has strong business and technical leadership. The company is doing all the right things at the right time, including infrastructure and process improvements, expansion, and preparation for future opportunities in quantum computing. Peter Chapman is a savvy business leader who understands what is essential to customers and is not afraid of tackling bold challenges. IonQ’s next announcement should be very interesting.