The Road to 1 Million Qubits: What Needs to Happen Next?

Scaling quantum computers to 1 million qubits is one of the most ambitious goals in modern quantum computing. It promises practical quantum advantage and fault-tolerant computation, but getting there requires solving major engineering, physical, and algorithmic challenges.

Here’s what needs to happen next, broken into key areas:

1. Error Correction & Fault Tolerance

• Current Problem: Qubits are fragile and prone to errors due to decoherence, crosstalk, and noise.
• Next Steps:
  • Implement quantum error correction (QEC) codes like surface codes that can detect and fix errors using logical qubits built from thousands of physical qubits.
  • Reduce the physical-to-logical qubit ratio, ideally below 1,000:1.
  • Improve gate fidelities and coherence times to support longer and more accurate computations.

2. Qubit Quality and Architecture

• Current Problem: Many qubit technologies (superconducting, trapped ions, neutral atoms, etc.) scale poorly or require extensive hardware overhead.
• Next Steps:
  • Develop modular, scalable architectures, such as:
    • Modular ion traps with photonic interconnects
    • Superconducting qubit tiles with on-chip control
    • Neutral atom arrays with reconfigurable topologies
  • Standardize inter-qubit connectivity, enabling flexible 2D or 3D architectures.
  • Improve qubit uniformity to avoid performance bottlenecks.

3. Cryogenic and Control Infrastructure

• Current Problem: Qubit systems (especially superconducting ones) require dilution refrigerators and massive classical control overhead.
• Next Steps:
  • Develop cryogenic CMOS control electronics to operate close to the qubits.
  • Create scalable I/O and wiring solutions to reduce the number of wires entering cryostats.
  • Integrate multiplexing and microwave control at scale to support millions of signals.

4. Fabrication and Integration

• Current Problem: No high-yield, wafer-scale quantum processor production exists yet.
• Next Steps:
  • Adopt semiconductor industry best practices for qubit fabrication.
  • Use 3D integration (e.g. chiplets, through-silicon vias) to stack classical and quantum components.
  • Ensure process repeatability and uniformity for mass production.

5. Software, Compilers, and Algorithms

• Current Problem: Even if hardware scales, software and algorithms must match to make use of 1 million qubits.
• Next Steps:
  • Develop fault-tolerant quantum compilers to optimize circuits for error-corrected hardware.
  • Build middleware for scheduling and error tracking across thousands of logical qubits.
  • Expand the portfolio of quantum algorithms that require high qubit counts (e.g., quantum simulation, machine learning, cryptography).

6. Interconnects and Networking

• Current Problem: Monolithic architectures are limited in size.
• Next Steps:
  • Build quantum interconnects (photonic links or entanglement distribution) to connect modules across a quantum network or quantum data center.
  • Lay groundwork for a quantum internet to scale beyond local processors.

7. Cost, Power, and Sustainability

• Current Problem: Scaling quantum systems currently requires large capital and energy-intensive infrastructure.
• Next Steps:
  • Innovate energy-efficient cryogenic systems.
  • Reduce cost per qubit through manufacturing and modularity.
  • Explore room-temperature qubit technologies (e.g., silicon spins, NV centers) as long-term alternatives.

Summary Roadmap to 1 Million Qubits