Guest Post By Izhar Medalsy, CEO and cofounder, Quantum Elements
and Prof. Daniel Lidar, CSO and cofounder, Quantum Elements
Most informed observers believe quantum is a transformative technology that will have a huge impact on people’s lives and likely play a pivotal role in solving many of the problems facing humanity today.
However, hardware progress in quantum computing has been uneven at best. Issues like coherence, noise, calibration, and scaling challenges, as well as the need for better error mitigation and correction, are slowing progress on the hardware side. Since there are different modalities and architectures, software must be recompiled for each platform. Consequently, software development lags as well. The industry needs a safe, fast and accurate environment to test hardware and algorithms before touching a real quantum device.
Most quantum software development teams still rely on classical simulators that break down as qubit counts rise. Enter digital twins. A staple of industries from aerospace to energy, digital twins are now being applied to quantum computing development with promising results and potentially huge implications.
A quantum digital twin is a physics informed software replica of a specific quantum device, not a generic simulator. It is essentially a virtual quantum machine running on a high-performance, classical computer, capturing all its behavior in real-time. For users, it’s the difference between a pilot in a flight simulator that can only allow for a straight flight in perfect weather conditions to a pilot having access to all the real-time realistic weather, airplane performance and conditions. Digital twins make it possible to develop and test quantum algorithms, control strategies, as well as error mitigation and correction techniques without depending exclusively on access to limited, fragile and costly quantum processors.
Quantum digital twins can also generate data used to train AI that learns how today’s quantum hardware behaves. In turn, these AI systems can help optimize hardware configurations and anticipate performance issues. They can capture all the functions of a quantum device and update as the hardware changes. The model always stays in sync with the physical machine.
Digital twins let end‑users prototype workloads without access to quantum hardware. Researchers can run thousands of experiments virtually before touching a quantum machine. Developers can test strategies on a model that behaves like real quantum devices. Manufacturers can explore error mitigation and correction, control‑pulse tuning, qubit layout changes, and noise‑source isolation in software rather than on a real machine not limited by the long hardware development timelines.
AI digital twins have already been proven practical and effective in replicating the operations of quantum computers on high performance classical ones. A team featuring researchers from AWS, USC, Harvard, and start-up Quantum Elements demonstrated a hardware-faithful digital twin capable of simulating a 97-qubit code with realistic noise in about an hour on a single AWS Hpc7a node. Using a new quantum Monte Carlo algorithm, they captured errors that traditional simulators miss. A normal simulation of 97-qubit code like this would require 497 entries, far beyond the capabilities of classical computers.
AI digital twins shorten the hardware learning loop from weeks to hours, reduce the cost of experimentation by orders of magnitude and democratize access. They create a space where hardware makers, software developers, and enterprise users can collaborate to build better machines. This paves the way for practical, noise-realistic digital-twins and faster progress towards fault-tolerant quantum computing.
The quantum industry must treat digital twins as an essential part of its infrastructure. Continuous‑learning AI digital twins will enable optimization of existing quantum devices and accelerate the path from NISQ to the future promise of quantum computing.
Bios:
Izhar Medalsy is a deep-tech leader with 15+ years of experience in advanced R&D, applied physics, and product development. He drives Quantum Elements’ mission to accelerate the transition from experimental quantum machines to scalable, production-grade systems. He holds a PhD in Physical Chemistry from the Hebrew University of Jerusalem and completed his postdoctoral research at ETH Zurich.
Daniel Lidar is the Viterbi Professor of Engineering at USC, where he holds appointments in Electrical & Computer Engineering, Physics, and Chemistry, and directs the Center for Quantum Information Science & Technology. He has been conducting research at the forefront of quantum innovation for nearly 30 years and has received numerous research awards, published hundreds of peer-reviewed papers, holds several patents, and has trained dozens of graduate students and postdocs.
Facts Only
Izhar Medalsy is CEO and cofounder of Quantum Elements and holds a PhD in Physical Chemistry from the Hebrew University of Jerusalem.
Daniel Lidar is CSO and cofounder of Quantum Elements, Viterbi Professor of Engineering at USC, and director of the Center for Quantum Information Science & Technology.
Quantum computing faces hardware challenges such as coherence, noise, calibration, and scaling issues.
Software development lags due to the need for recompilation across different quantum architectures.
Digital twins are physics-informed software replicas of specific quantum devices, not generic simulators.
A quantum digital twin runs on high-performance classical computers and captures real-time behavior of quantum machines.
Researchers from AWS, USC, Harvard, and Quantum Elements demonstrated a digital twin simulating a 97-qubit code with realistic noise in about an hour on a single AWS Hpc7a node.
Traditional simulators would require 497 entries to simulate a 97-qubit code, which is beyond classical computing capabilities.
Digital twins enable testing of quantum algorithms, error mitigation, and hardware optimizations without relying on physical quantum processors.
AI can use digital twin-generated data to optimize quantum hardware configurations and predict performance issues.
Digital twins reduce experimentation costs, shorten development timelines, and improve collaboration among hardware makers, software developers, and enterprise users.
The quantum industry is encouraged to adopt digital twins as essential infrastructure to accelerate progress toward fault-tolerant quantum computing.
Executive Summary
Full Take
The narrative presents digital twins as a breakthrough solution for quantum computing’s hardware and software bottlenecks, and the strongest version of this argument is compelling. Digital twins address critical pain points—costly experimentation, limited access to quantum processors, and the inefficiency of generic simulators—by providing a high-fidelity, physics-informed virtual environment. The demonstrated 97-qubit simulation with realistic noise is a significant achievement, showcasing the potential to bridge the gap between NISQ devices and fault-tolerant systems. The authors rightly highlight the collaborative benefits, where hardware makers, developers, and end-users can iterate rapidly without the constraints of physical hardware.
However, the discussion leans heavily on the promise of digital twins without deeply interrogating their limitations. For instance, how accurately can digital twins replicate the full spectrum of quantum noise and error sources? The article assumes that AI-trained on digital twin data will reliably optimize hardware, but real-world quantum systems may exhibit unpredictabilities not captured in simulations. Additionally, the claim that digital twins "democratize access" presupposes widespread availability of high-performance classical computing resources, which may not be equally accessible to all researchers or institutions.
The root cause of this narrative is the tension between quantum computing’s vast potential and its current practical limitations. The industry’s urgency to deliver on its promises creates pressure to find workarounds, and digital twins emerge as a plausible shortcut. Yet, this approach risks becoming a crutch—if digital twins are treated as a substitute for fundamental hardware advancements rather than a complement, progress toward fault tolerance could stall. Historically, similar patterns have emerged in other tech sectors where simulation tools delayed necessary breakthroughs in physical systems.
For human agency, digital twins could lower barriers to entry, enabling more diverse participation in quantum research. However, if access to these tools remains concentrated among well-funded institutions, they may exacerbate existing disparities. Second-order consequences include the potential for over-reliance on simulations, leading to algorithms that perform well in virtual environments but fail in real-world quantum hardware.
Bridge questions: What evidence would demonstrate that digital twins can accurately predict real quantum hardware behavior beyond current benchmarks? How might the reliance on digital twins shift the balance of power in quantum research, and who stands to gain or lose from this shift? If digital twins become the primary testing ground, what safeguards are needed to ensure they don’t become a bottleneck for innovation?
Counterstrike scan: A coordinated influence campaign pushing this narrative might emphasize digital twins as a silver bullet, downplaying their limitations while framing skeptics as obstructionists. The actual content does not match this pattern—it acknowledges challenges and presents digital twins as a tool rather than a panacea. The authors’ credentials and the inclusion of a concrete demonstration lend credibility without overpromising.