Welcome to issue #21 of This Week in Quantum — a roundup of the most important developments in quantum computing from July 27 to August 2, 2026.
Quantum computing’s public-market moment continued this week as D-Wave joined the growing list of quantum companies trading on Nasdaq. On the research side, two Nature papers pushed error-corrected hardware forward, while a quieter but significant story broke in cryptography: an AI model helped find a flaw in a post-quantum signature algorithm before it could be standardized — arguably the system working exactly as intended.
Industry News
D-Wave rings the Nasdaq opening bell
On July 27, D-Wave Quantum’s CEO Dr. Alan Baratz and the company’s management team rang the Nasdaq Opening Bell at the Nasdaq MarketSite in Times Square, marking D-Wave’s listing debut on the exchange following a voluntary transfer of its stock listing. The company, which describes itself as the only dual-platform provider of both annealing and gate-model quantum systems, framed the move as recognition of its commercial traction, citing its Leap cloud service and more than 100 customer organizations across commercial, government, and research sectors.
Infleqtion names a new SVP of Quantum Computing Systems
On July 28, Infleqtion announced the appointment of Dr. Joseph Buck, a Lockheed Martin Senior Fellow and Caltech-trained neutral-atom physicist, as Senior Vice President of Quantum Computing Systems, adding senior technical leadership as the company continues to expand its government and commercial contract base.
Research Highlights
Microsoft and Quantinuum report error-correction gains of up to 800x
A Nature study from Microsoft Quantum and Quantinuum, highlighted this week, showed that quantum error-correction techniques reduced computational errors by factors ranging from 11x to 800x compared with running the same calculations directly on physical qubits. Using two error-correction code constructions — a 12-qubit code encoding two logical qubits and a 16-qubit tesseract color code encoding four — combined with error detection and post-selection on a trapped-ion QCCD processor, the team demonstrated repeated mid-computation error correction across experiments involving up to 12 logical qubits. The results suggest today’s processors can already benefit meaningfully from fault tolerance, though further advances in hardware and real-time decoding are still needed at larger scale.
HRL Laboratories demonstrates a self-running silicon QPU
Also published in Nature this week, HRL Laboratories detailed an 18-qubit silicon spin quantum processing unit that operates autonomously at 4 Kelvin using a custom cryogenic CMOS control chip. Fabricated on standard silicon-germanium wafers, the integrated architecture significantly reduces error rates and demonstrates fault-tolerant functionality without relying on room-temperature control electronics — a step toward more compact, scalable silicon-based systems.
Security Watch
An AI model helps catch a flaw in a candidate post-quantum algorithm
On July 28, Anthropic announced it had used an AI model to discover a vulnerability in HAWK, a lattice-based digital signature algorithm that was under consideration for standardization. The HAWK development team withdrew the algorithm from consideration as a result, and it will not be standardized or deployed. NIST confirmed the finding does not affect any of its already-finalized PQC standards, including ML-KEM and ML-DSA, which rely on different mathematical foundations and remain secure for implementation today. The episode is a useful reminder that the PQC standardization process is still actively vetting candidate algorithms, and that catching a flaw before deployment is the process working as designed.
The Bigger Picture
This week paired two threads that define where quantum computing is right now: hardware inching toward genuine fault tolerance in the lab, and public markets treating the sector as investable infrastructure rather than a research bet. The HAWK withdrawal adds a third thread worth watching — cryptographic standardization is not a one-time event but an ongoing, adversarial process, and AI-assisted cryptanalysis is becoming part of how that process gets stress-tested before deployment rather than after.
Upcoming Events
- September 9-11 — Quantum Expo Korea 2026, Seoul
- September 13-18 — IEEE Quantum Week, Toronto
- September 28-30 — Quantum industry conference, Dubai