The IonQ Photonic Interconnect Milestone That DARPA Just Backed — and What It Means for Networked Quantum Systems — and What It Means for Networked Quantum Systems

The air in a small lab in College Park, Maryland, hums with the kind of quiet only laser systems and refrigeration units can create. It’s the kind of location that doesn’t appear to be historic from the outside. There are server racks, white walls, and a few engineers carrying coffee cups.

However, something happened there recently that, depending on who you ask, either matters a great deal or is being subtly overhyped by an industry in dire need of positive news.

This week, IonQ declared that two separate trapped-ion quantum computers have been photonically connected. To put it simply, the company was able to entangle two different machines by sending single photons between them. According to the company, this is the first time that commercial systems have been used instead of meticulously adjusted lab prototypes. It matters that there is a difference. While commercial machines must operate on a Tuesday morning with the HVAC system running, lab demonstrations take place in controlled environments.

The timing is what makes the announcement more difficult to accept. IonQ secured a position in DARPA’s HARQ program earlier that week, which aims to integrate various qubit types into networked architectures. When combined, the two announcements seem more like choreography than coincidence.

Defense organizations typically fund things they have already seen demonstrated, even covertly, rather than things that might work in the future. Behind the press releases, that provides some insight into the true state of this technology.

The CEO of IonQ, Niccolo de Masi, referred to the achievement as “pivotal”—a term that has grown a little stale in tech communications but may be justified in this instance. It is structural in nature. It is not possible to simply enlarge quantum computers in the same manner as classical computers. You can’t expect things to scale gracefully by simply adding more qubits to a single chip. Coherence breaks down, mistakes pile up, and the entire architecture starts to wobble. The majority of physicists now concur that modularity—many smaller machines connected by light—is the solution. IonQ is placing that wager, and it now has a functional demonstration to support it.

Compared to the quantum hype cycles of even three or four years ago, it is difficult to ignore how different this moment feels. Qubit counts dominated discussions back then, as if quantum computing were a horsepower competition. Fidelity, networking, and error correction—less glamorous metrics that really determine whether any of this becomes useful—are now the focus of the discussion. The same pattern can be seen in IonQ’s earlier this year achievement of 99.99% two-qubit gate fidelity. The business appears to be quietly accumulating the unsexy victories.

However, there are many reasons to be dubious. IonQ is neither close nor profitable. Over the past year, its stock has fluctuated greatly, driven in part by real technological advancements and in part by the same speculative energy that has flooded into anything labeled “AI” or “quantum.” By 2028, some analysts predict revenues will be close to $390 million, which would necessitate rapid expansion. Some believe the sector as a whole is still years away from having a significant commercial impact, while others see a more modest path.

None of that is resolved by the interconnect milestone. Maybe it shifts the question a little. In the past, the question of whether networked quantum systems could be produced on a commercial scale was debatable. The argument is beginning to wane. The new question, which is messier and more intriguing, is whether IonQ or anyone else can produce them quickly, affordably, and consistently enough to be significant. As this develops, it seems as though technology is finally surpassing marketing. It’s another matter entirely if the company succeeds.