Quantum Computing News: IBM Gets $1B Award as IONQ Taps Nvidia AI

Welcome to Thursday's quantum computing news recap. This edition looks at five developments across the sector, with International Business Machines Corporation (IBM) backing a new $1 billion U.S. quantum foundry award, IQM Quantum Computers (IQMX) expanding into South America, IonQ, Inc. (IONQ) and Nvidia Corporation (NVDA) testing generative AI for quantum optimization, D-Wave Quantum Inc. (QBTS) preparing new annealing and gate-model demonstrations, and Oracle Corporation (ORCL) adding post-quantum cryptography support to Java 27.

1. IBM's Anderon Finalizes $1 Billion U.S. Quantum Foundry Award

Anderon, an IBM company, finalized a $1 billion CHIPS and Science Act award with the U.S. Department of Commerce to support quantum wafer manufacturing R&D at its 300-millimeter foundry in Albany, New York. IBM is separately investing another $1 billion in the venture, which has already begun running its first quantum wafers for superconducting qubit arrays, quantum input/output signaling, and readout components.

The scale of the combined funding is of significance because manufacturing remains one of the biggest constraints on expanding quantum hardware, and the next test will be whether Anderon can move from early wafer runs into reliable production for IBM and outside customers.

2. IQM Lands First South American Quantum Computer Deal

IQM Quantum Computers signed a purchase agreement with Brazil's Eldorado Research Institute for its first quantum computer in South America, with an IQM Spark system scheduled for installation in Campinas in Q1 2027. The deal also gives Brazilian researchers remote access to IQM's 54-qubit cloud system in Europe, allowing teams to develop applications locally before moving larger workloads to higher-capacity hardware. Financial terms were not disclosed, but the order adds to IQM's expansion after the company reported a €102 million backlog in August, making new system deliveries and backlog conversion the main numbers to watch.

3. IonQ and Nvidia Use Generative AI to Cut Quantum Circuit Tuning Time

IonQ, Inc., Nvidia Corporation, Oak Ridge National Laboratory, and the University of Tennessee reported a generative AI method that can create quantum optimization circuits directly instead of relying on repeated parameter tuning. In benchmark tests, circuit-generation time stayed near 28 seconds across the tested problem sizes, while the previous method rose from about 34 seconds at four qubits to more than 11 minutes at 12 qubits, with solution quality roughly doubling as subproblem size increased on a 100-variable benchmark.

The result could help reduce one bottleneck in hybrid quantum optimization, but the work was performed on an NVIDIA H200 GPU simulation rather than quantum hardware, so execution on real IonQ systems is the more important next validation step.

4. D-Wave Prepares Dual-Platform Quantum Push at World Congress

D-Wave Quantum Inc. plans to showcase both its annealing and gate-model systems at Quantum World Congress from September 23 to 25, highlighting a strategy that combines near-term optimization workloads with longer-term fault-tolerant quantum computing. The company is expected to present commercial and public-sector annealing results alongside updates on its dual-rail gate-model architecture, which it says could reduce the resources needed for quantum error correction.

D-Wave has added gate-model capability through its $550 million acquisition of Quantum Circuits Inc., secured a $1.5 million NSF grant for related work, and received up to $100 million through the CHIPS R&D program. Investors will be watching for progress toward D-Wave's planned first dual-rail gate-model machine later this year.

5. Oracle Adds Post-Quantum Protection to Java 27

Oracle Corporation released Java 27 with new post-quantum cryptography features aimed at protecting enterprise applications from future quantum attacks. The update adds hybrid TLS 1.3 key exchange through standard Java security interfaces and includes Oracle Jipher 20, which supports NIST-standardized ML-KEM and ML-DSA algorithms in regulated environments. The change is more relevant to enterprise security adoption than quantum hardware demand, but it shows post-quantum cryptography moving deeper into widely used software infrastructure. The next point to watch is how quickly similar protections are backported into long-term support Java releases used across large enterprise systems.