Quantum ComputingTimeline Predictions
Every major quantum company predicts a different date for the big milestones. Here's when each expects quantum computers to deliver scientific breakthroughs, break modern encryption, and reach widespread commercial impact.
Company/Organization | Isolated Scientific Breakthrough (probably in physics) | Break Encryption (RSA 2048) | Widespread Impact (probably in materials science) | Source |
|---|---|---|---|---|
 Atom Computing Scale neutral atoms into thousands of qubits for commercial use. | ~2026 | — | — | View Source |
D D-Wave High-coherence dual-rail superconducting gate-model qubits that detect ~90% of errors, targeting a 10-logical-qubit fault-tolerant system in 2030 and 100 logical qubits by 2032. | ~2030 | — | ~2032 | View Source |
 DARPA Scale quantum computers to 1000 qubits by 2028. | — | — | ~2033 | View Source |
D DOE Build and deploy the first scientifically relevant, fault-tolerant quantum computer for research by 2028. | ~2028 | — | — | View Source |
 Fujitsu Build 10,000+ qubit superconducting machine by 2030, with fault-tolerant design. | — | — | ~2035 | View Source |
 Google Pursue both superconducting qubits and neutral-atom systems to reach commercially useful quantum computing by the end of the decade. | ~2025 | ~2029 | ~2030 | View Source |
 IBM Grow superconducting qubits with modular chips and error correction to reach fault-tolerant systems. | ~2026 | — | ~2029 | View Source |
 IonQ Scale trapped-ion systems with photonic links and chip miniaturization. | ~2025 | ~2030 | ~2030 | View Source |
 Jensen Huang (NVIDIA) Quantum is still over 15–30 years away; GPUs will dominate until then. | — | — | <2045 | View Source |
 Microsoft Topological qubits on a lead-based materials stack. After a ~1,000× reliability gain with Majorana 2, Microsoft says it has cut its timeline in half and now aims for a scalable quantum computer by 2029. | ~2029 | — | ~2035 | View Source |
M Mikhail Lukin (Harvard) Neutral-atom error-correction breakthroughs are pulling fault-tolerant timelines forward. | ~2026 | — | <2030 | View Source |
 NIST Mandate shift to quantum-safe cryptography by 2030–2035. | — | ~2035 | — | View Source |
 Oratomic Use optical-tweezer neutral atoms to reach utility-scale, fault-tolerant quantum computing by the end of the decade. | ~2027 | — | ~2030 | View Source |
 PsiQuantum Build million-qubit photonic computers in data centers. | — | ~2030 | ~2029 | View Source |
 Quantinuum Use trapped ions with high fidelity to reach universal fault tolerance by 2030. | ~2025 | — | ~2030 | View Source |
Q QuEra Neutral-atom systems scaling to fault tolerance on the cloud: a megaquop-class Libra machine in 2028, then a gigaquop-class system with 1,000+ logical qubits around 2028–2029. | ~2028 | — | ~2029 | View Source |
 Rigetti Build modular superconducting systems, scaling to 1000+ qubits later this decade. | ~2025 | — | ~2029 | View Source |
 Vitalik Buterin Warns crypto must prepare for quantum threats sooner than many expect. | — | 20% chance of 2030 | — | View Source |
Timeline Categories
Isolated Scientific Breakthrough
(probably in physics)
When quantum computers will solve meaningful scientific problems faster than classical computers
Break Encryption
(RSA 2048)
When quantum computers can factor large numbers fast enough to break RSA 2048-bit encryption
Widespread Impact
(probably in materials science)
When quantum computing becomes commercially viable and widely adopted across industries
Best guess: Generating chemistry and materials data to train AI models. Many companies are now focusing on this.
Other Application Areas: Timeline Unclear
These areas showed early promise, but the path forward is less certain than initially hoped.
Optimization
The promise: Revolutionize logistics, finance, and supply chains.
Current status: The quantum advantage for optimization requires much larger systems than near-term machines. Classical algorithms continue to improve rapidly, making the target harder to reach.
Machine Learning
The promise: Exponential speedups for AI and neural networks.
Current status: Most ML tasks are better suited to classical hardware like GPUs and TPUs. The data loading bottleneck and specific quantum constraints limit practical advantages for typical ML workflows.
Genomics
The promise: Quantum pattern recognition could advance personalized medicine.
Current status: Genomic analysis primarily involves processing large datasets rather than quantum mechanical problems. Classical big data tools and AI are proving more practical for these data-intensive tasks.
Key insight: The clearest timelines exist for well-defined mathematical problems: factoring (for breaking encryption) and simulating molecules. Other applications have fuzzier targets.
Key Takeaway
The consensus seems to be 2030-2035 to break RSA-2048. Commercial applications are further out.