Quantum computing has captivated scientists and industry leaders alike with its promise to tackle problems far beyond the reach of classical machines. Erik Hosler, a photonics strategist steering scalable device research at PsiQuantum, acknowledges that moving from proof-of-concept experiments to business-grade systems depends on one decisive factor. Scale. Without enough qubits and without the ability to run long, reliable sequences of operations, quantum computers remain confined to the lab.
The excitement around early quantum prototypes often obscures the true challenge. While a handful of entangled qubits may highlight exotic phenomena, they can’t deliver solutions to real-world problems. To cross the threshold from demonstration to disruption, quantum systems must grow in both size and stability, transforming scientific curiosity into commercial capability.
Why Qubit Quantity Matters More Than Hype
Early successes generated headlines, but they also exposed a stark reality. More qubits don’t always mean more value. When error rates climb with each additional qubit, computational fidelity collapses. In practice, useful applications require systems that can sustain complex calculations, not just fleeting quantum states.
The industry consensus now points to a clear target: hundreds to thousands of usable qubits. At that scale, quantum devices could tackle chemistry simulations, advanced optimization problems, and machine-learning tasks that stump classical supercomputers. Getting there, however, demands overcoming noise, crosstalk, and thermal fluctuations, challenges that multiply as qubit counts rise.
Physical vs. Logical Qubits: The Hidden Overhead
A critical distinction lies between physical qubits, the raw quantum elements, and logical qubits, which are error-protected constructs formed by grouping many physical qubits. Physical qubits alone are highly susceptible to decoherence; stray electromagnetic fields, minute temperature shifts, and even cosmic rays can corrupt their state. Error correction schemes mitigate these effects by continuously monitoring qubit health and applying corrective operations, but at a steep cost.
Current estimates suggest that creating one logical qubit may require hundreds of physical qubits. Thus, a system with one million physical qubits might yield only a few thousand logical qubits, barely enough to run the depth of circuits needed for meaningful tasks. This overhead underscores why raw qubit counts can be misleading. What truly matters is the number of logical qubits capable of supporting long computational sequences.
Scaling Error Correction for Billions of Operations
Quantum algorithms of commercial interest often involve billions of sequential operations. For example, simulating molecular interactions can require deep circuits with thousands of multi-quit gates. Each gate must execute with high fidelity to prevent error accumulation. Error-correction routines, however, introduce additional gates and measurements, increasing the circuit depth further. Erik Hosler notes,
“We need hundreds to thousands of usable qubits with the capability to do billions of sequential operations to really do useful work.”
This benchmark captures two interlinked demands, qubit scale and operational depth. It implies that even if a system reaches the necessary logical qubit count, it must also maintain coherence across extensive gate sequences. Achieving both simultaneously remains the field’s most formidable technical barrier.
Engineering Coherence: Materials, Shielding, and Cryogenics
Maintaining quantum coherence is akin to keeping a soap bubble intact in a hurricane. Any disturbance, vibration, stray photon, or temperature drift can cause the qubit’s quantum state to collapse. To counteract this, quantum processors operate in mK-level cryostats, isolated from external noise by layers of magnetic shielding and vibration-damping mounts.
Materials science plays a pivotal role. Superconducting qubits, for instance, rely on ultra-pure aluminum or niobium films to minimize energy loss. Photonic qubits, encoded in single photons, require waveguides with minimal scattering and low-loss optical components. Advanced packaging techniques, such as three-dimensional integration and hermetic sealing, further reduce interference and enable tighter qubit spacing without sacrificing isolation.
Mass Manufacturing: Adapting Semiconductor Foundries
Translating lab-scale qubit fabrication into mass production demands rethinking semiconductor workflows. Traditional fabs excel at crafting billions of transistors on wafers, but quantum devices introduce new elements like Josephson junctions, photonic waveguide networks, or ion-trap electrodes. Each requires specialized process steps, like electron-beam lithography for sub-50-nm features or cryo-compatible packaging for on-chip refrigeration.
Photonics-based approaches offer a bridge by leveraging silicon-photonics platforms; qubit devices can ride on the coattails of mature semiconductor infrastructure. Yet adapting these platforms to quantum specifications means tightening overlay tolerances, improving line-edge roughness, and integrating optical components without degrading yield. Achieving high yield at scale will be key to bringing qubit counts into the hundreds of thousands that error-corrected systems demand.
Control Electronics: Orchestrating Thousands of Channels
Each qubit requires precise control signals, microwave pulses, laser beams, or voltage ramps to perform logic gates. Scaling to thousands of qubits demands hundreds or thousands of control channels, all synchronized with sub-nanosecond timing accuracy. Conventional room-temperature electronics face bandwidth and latency limits when interfacing with cryogenic hardware.
Researchers are exploring cryo-CMOS control chips placed directly inside dilution refrigerators, reducing signal path length and improving timing precision. On-chip Digital-to-Analog Converters (DACs) and Analog-to-Digital Converters (ADCs) can manage multiple qubits per channel, but they must operate reliably at mK temperatures. Innovations in multiplexing, custom FPGA architectures and low-noise amplifier designs will be crucial to scale control hardware alongside qubit arrays.
Software and Algorithmic Innovations
Hardware advances only unlock value when paired with software that can exploit them. Noise-aware compilers, for instance, map high-level quantum circuits onto physical qubits by minimizing error-prone operations and choosing more reliable gate sequences. Quantum-inspired classical algorithms can tackle portions of a problem, reducing quantum circuit depth.
Frameworks like Qiskit, Cirq, and PennyLane already provide libraries for error correction and circuit optimization. As qubit counts rise, these tools must develop to manage resource allocation across thousands of qubits and billions of operations. Automated calibration routines, adaptive error-mitigation protocols, and real-time feedback loops will become standard components of quantum software stacks.
The Road Ahead: Incremental Milestones to Mega-Scale
Reaching the era of millions of physical qubits and thousands of logical qubits will not happen overnight. Instead, the industry will advance through incremental milestones:
- Improved Physical Qubit Coherence: Pushing coherence times from microseconds to milliseconds through materials and cooling breakthroughs.
- Optimized Error Correction: Demonstrating fault-tolerant logical qubits at the tens-to-hundreds scale, reducing physical-to-logical qubit overhead.
- Integrated Control Electronics: Deploying cryogenic control chips in multi-module systems to streamline signal routing and timing.
- Foundry-Level Fabrication: Establishing pilot lines for quantum device production with yields above 90%.
- Hybrid System Prototypes: Combining different qubit types and interconnects into cohesive platforms capable of running mid-scale applications reliably.
Each milestone will build confidence, attract investment, and deliver small pockets of real-world value, laying the groundwork for the quantum computers of tomorrow.
Scaling as the Gateway to Impact
The promise of quantum computing rests on more than exotic physics; it requires a relentless focus on engineering scale, coherence, and integration. Qubit count alone is insufficient; what matters is the number of logical qubits that can sustain billions of operations without succumbing to errors. Overcoming this barrier will demand advances across materials science, fabrication technology, control electronics, and software.
Meeting the dual challenges of scale and stability is the linchpin for quantum commercial success. Once quantum systems achieve the required qubit scale, they will unlock vital applications in chemistry, logistics, finance, and beyond, ushering in the long-awaited quantum payoff.