A U.S. chip factory buyout, $25.5M in early stage funding for a new quantum chip design, and the first public U.S.
This week a fiber link between Evanston and downtown Chicago carried two kinds of signal at the same time: 1.6 Tbps of normal internet traffic and entangled photons that held 94.2 percent of the standard yardstick for how quantum a connection really is. The photons rode 15.2 miles of cable already in commercial use, separated by wavelength filters on the O-band, the slice of the spectrum telecom fibers already dedicate to data. Northwestern researchers built the link and reported the figure in Optica Quantum, with Tech Times and The Quantum Insider echoing the result.
Quantum links have long needed dark fiber because ordinary light swamps single-photon signals. Filtering quantum photons into the same band telecom operators already use for dense data traffic means a quantum testbed can ride on infrastructure that exists, not on cable a lab has to install. The Chicago run is the first published instance of entangled photons surviving that coexistence at telecom-class classical bitrates.
A second coexistence result landed the same week. Qunnect switched on ABQ-Net in Albuquerque, an open-access test network in the U.S. that lets companies try quantum links without building their own fiber. Infleqtion, Aliro, Tensora, and Bandelier Technologies have signed on to use the network for timing synchronization, key generation, and geosensing experiments. ABQ-Net is a testbed, not a production quantum internet, and Qunnect built it on existing fiber. That puts the second U.S. entanglement link of the week on the same "share the cable" pattern as the first.
A third thread sat in the background. Researchers at Wits University reported topology-protected quantum information in light that survives atmospheric turbulence, a result that points toward weather-resilient free-space links rather than coexistence with telecom traffic. The Wits work is a research milestone, not a deployable network, and it rounds out the week's pattern: photons holding their quantum state under realistic conditions, even if those conditions differ from O-band sharing.
Capital moved in a second direction. ZuriQ closed a $25.5 million seed round led by Quantonation to scale a 2D trapped-ion architecture, a geometry that escapes the linear chain of ions most labs use today, that the company says uses Penning micro-traps and static magnetic fields. ZuriQ has a 3-by-3 demonstrator; the seed targets hundreds of qubits on a single chip, a stage the chip hasn't reached. Hangzhou-based Heguang Quantum reported a separate seed and progress on deterministic photonic GKP states, error-protected quantum states built into light on demand, through a pluggable nonlinear module, with an HQ10 hybrid accelerator targeted for 2027. Both bets sit outside the superconducting and linear-chain ion paths that dominate the current quantum hardware map.
The week's third thread was consolidation. IonQ cleared its final regulatory hurdle to acquire SkyWater Technology, with closing set for July 31. SkyWater runs a domestic foundry, and the deal makes IonQ a vertically integrated full-stack quantum company with its own U.S. chip capacity. In the same week, AT&T signed an agreement to expand D-Wave quantum annealing across network operations, an announcement that describes an expansion plan rather than a deployed rollout.
Three threads, one direction. Quantum demonstrations are starting to run on shared telecom cables, in shared fabs, and with the same kind of early-stage capital gravity that earlier platform shifts saw. None of it is a quantum internet yet. The Chicago run is a 15-mile link, ABQ-Net is a test network, and the ZuriQ and Heguang chips are still seed-stage. The watch item for the next quarter is whether the O-band coexistence result holds up in independent replications and whether IonQ's foundry capacity changes what the rest of the field can buy.