Imagine a memorandum of understanding that knit together PsiQuantum’s photonic, manufacturable route to fault tolerance with Lockheed Martin’s warfighting systems engineering and platform scale. That kind of document would not be a simple vendor contract. It would be a statement of intent about how the United States prepares for a world where error corrected quantum machines tilt the balance on simulation, optimization, and sensing.
The scenario is not fanciful. Over the past year PsiQuantum has been moving from lab demonstrations toward production scale. The company has advanced in DARPA’s benchmarking and validation pipeline for utility scale quantum computing, an endorsement of their photonic architecture and road map toward a fault tolerant design. PsiQuantum has also signed defense-focused contracts to deliver photonic chip capabilities to the Air Force Research Laboratory, work that explicitly ties their manufacturable Omega chipset and electro-optic components to military research use cases. And independent reporting has documented PsiQuantum’s push to industrialize chip fabrication in partnership with GlobalFoundries, a practical step that changes timescales for fieldable systems.
Lockheed Martin, for its part, is already building quantum pedigree in sensing and systems integration. The company leads a DIU-backed prototype effort to get quantum-enabled inertial navigation out of the lab and onto platforms that must operate without GPS. That program highlights Lockheed’s appetite for maturing quantum sensors into rugged, mission-ready hardware.
Put those threads together and the contours of a useful MoU emerge. It would likely include three parallel tracks. First, algorithm and application co-design. Lockheed would present prioritized mission problems where classical compute struggles: high fidelity fluid dynamics and multiphysics for propulsion and airframe design, quantum chemistry for new materials and energetics, and combinatorial optimization for logistics and mission planning. PsiQuantum would commit engineering cycles and access to its Construct software stack and test circuits so Lockheed teams could co-develop fault-tolerant algorithms and early hybrid workflows that map to photonic hardware characteristics. The DARPA validation work PsiQuantum has been through shows why that co-design is urgent: the architecture tradeoffs made today will lock in which applications scale efficiently tomorrow.
Second, hardware transition and resilience workstreams. Defense adoption demands reproducible, hardened hardware. PsiQuantum’s move toward semiconductor fab production and its BTO electro-optic components create a pathway to repeatable modules that can be qualified for defense environments. An MoU would probably carve out joint road maps for supply chain resilience, environmental hardening, and shared testbeds where Lockheed could validate components under vibration, radiation, and lifecycle stresses common to aerospace systems. The AFRL chip delivery contract suggests PsiQuantum already has the mechanisms to supply and iterate on designs with military labs.
Third, secure integration and policy guardrails. Quantum systems present dual use questions: breakthroughs in algorithmic simulation or code breaking have immediate national security implications. Any serious MoU would therefore bind both parties to clear export control, cybersecurity, and classification practices while establishing governance for how prototype results map back to acquisition pathways. Lockheed’s role as an integrator that routinely handles classified programs would be essential for translating lab-scale capability into fieldable, policy-compliant systems.
There are also strategic tempo considerations. If PsiQuantum’s path to utility-scale, error-corrected machines continues to look feasible inside a decade, then investing now in mission-relevant algorithms and integration testbeds buys time. Hard problems like chemical kinetics for fuels or turbulent flow modeling will not become solvable overnight just because a million-qubit machine exists. But pre-built algorithm families, validated on hybrid classical-quantum pipelines and mapped to photonic connectivity graphs, would be ready the moment hardware crosses the fault tolerance threshold. DARPA’s benchmarking program is aimed precisely at that readiness acceleration.
Finally, expect friction. The defense sector values provenance, reproducibility, and long product lifecycles. Startups move faster and prioritize different KPIs. A credible MoU would need explicit milestones, escrowed design artifacts, and perhaps even joint manufacturing commitments to bridge that cultural divide. It would also have to address the geopolitical angle: when quantum advantage becomes a strategic asset, shielding critical intellectual property and ensuring allied access will be as important as the science itself.
We are living through a transition from quantum promise to quantum engineering. The most consequential MoUs will be those that do more than buy laboratory hours. They will institutionalize shared road maps for applications, hardware transition, and governance. For national security that means not only asking whether a quantum computer can solve a problem, but also whether it can be trusted, supplied, and integrated into systems that operate in contested environments.
If and when a formal understanding ties a scale-capable vendor to an integrator with Lockheed’s reach, policymakers and technologists should watch three signals closely: concrete co-developed algorithms for defense-relevant physics and optimization, demonstrable component qualification work in military testbeds, and binding agreements on supply chain and information governance. Those are the things that turn a speculative advantage into an operational one.
Quantum will not replace the fundamentals of deterrence and diplomacy. But an MoU that organizes science, engineering, and policy around fieldable quantum capabilities would change the calculus for how fast advanced simulation, sensing, and optimization make it from white papers to war rooms.