Why 2025–2026 Marks a Genuine Inflection Point
This article is written for technically literate readers who want more than headlines. It assumes familiarity with the broad concept of quantum computing but defines every specialised term on first use. The aim is a calibrated assessment: neither dismissing China’s progress as mere state propaganda nor amplifying it into an imminent strategic catastrophe.
Quantum computing has been accompanied by a persistent hype cycle since at least 2019. Announcements of “quantum supremacy,” record qubit counts, and unprecedented computational speeds have followed one another at pace — and most have been accompanied by significant caveats that the headline coverage glosses over. Understanding why 2025–2026 is different from previous cycles requires a clear framework for what a quantum milestone actually measures.
From lab demonstrations to commercial deployments: a shift in maturity
Until 2023, virtually all significant quantum demonstrations were one-off academic experiments — devices assembled for a specific benchmark, run once, and then described in a paper. What distinguishes 2025–2026 is a visible shift toward repeated, accessible, and in some cases commercially deployed systems. China’s Hanyuan-1, discussed in detail below, represents a 100-qubit neutral-atom machine offered through cloud access — not a laboratory curiosity. Origin Quantum’s Tianyan-504 superconducting processor is similarly framed around commercial and research cloud services rather than isolated experiments.
This shift in maturity matters because it changes the nature of the evidence. Cloud-accessible systems can be independently benchmarked by third parties; their performance claims are, in principle, falsifiable in a way that single-lab demonstrations are not. That does not mean every commercial claim is verified — it means the evidentiary standard has improved.
How to read quantum milestones without the hype
Three distinctions are essential throughout this article and are worth establishing upfront.
First: physical qubit count is not the same as capability. A physical qubit is the raw hardware unit — a superconducting circuit, a trapped ion, a photon — that carries quantum information. Physical qubits are noisy; they make errors. The number of physical qubits in a device tells you about scale, but almost nothing about what useful computation it can perform.
Second: quantum advantage — the point at which a quantum computer solves a problem faster or more efficiently than any classical computer — is task-specific and contested. It is not a binary threshold. Quantum supremacy, the stronger claim that a quantum device has performed any computation beyond classical reach, has been asserted by Google (2019), China’s Jiuzhang team (2020, 2023), and others. Those claims are real but narrow: the computations involved have no direct practical application.
Third, the concept that animates most security discussion — a cryptographically relevant quantum computer, or CRQC — describes a machine capable of running Shor’s algorithm at sufficient scale to break widely deployed public-key encryption. No such machine exists today. The question this article addresses is how close China, and the field generally, is to building one.
With that framework in place, the hardware evidence can be examined on its own terms.
The Zuchongzhi Lineage: From 1.0 to 3.0 and Beyond
The Zuchongzhi series, developed by a team at the University of Science and Technology of China (USTC) led by Jian-Wei Pan, is China’s flagship superconducting quantum processor programme. Superconducting qubits — tiny circuits cooled to near absolute zero that exploit quantum mechanical properties of superconducting materials — are the same hardware architecture used by Google and IBM, making cross-platform comparison more straightforward than with exotic modalities.
Zuchongzhi 3.0: 105 qubits and what that number actually measures
In March 2025, the Chinese government announced Zuchongzhi 3.0 as a 105-physical-qubit superconducting processor (english.www.gov.cn, March 2025). The qubit count alone would be unremarkable — IBM’s Eagle processor reached 127 qubits in 2021. What matters is what the team claims alongside that number: a quantum error rate and, crucially, a demonstration of below-threshold error correction.
Qubit fidelity — the accuracy with which a qubit operation is performed — is typically expressed as a gate error rate. For Zuchongzhi 3.0, the USTC team reported single-qubit gate fidelities above 99.9% and two-qubit gate fidelities in the 99.7–99.8% range. These figures are competitive with the best published results from Google and IBM, though independent verification through peer-reviewed publication is the appropriate standard rather than government announcement alone.
Below-threshold error correction: why this matters more than qubit count
Error correction is the central unsolved engineering challenge in quantum computing. Because physical qubits are inherently noisy, useful large-scale computation requires encoding information into logical qubits — protected units built from many physical qubits working together to detect and correct errors. The “threshold” refers to the error correction threshold: the maximum physical error rate below which error correction schemes actually improve reliability rather than making things worse.
Demonstrating below-threshold operation — that is, showing that adding more error-correcting qubits does in fact suppress the logical error rate — is a qualitative milestone. It is the difference between a device that gets noisier as it scales and one that could, in principle, be scaled toward fault tolerance. Google’s Willow chip (December 2024) made the same claim and published it in Nature; Zuchongzhi 3.0’s claim places China in the same experimental class.
This is the content gap in most coverage: journalists report the qubit number; the technically significant result is the error correction behaviour. A 105-qubit device demonstrating below-threshold error correction is a more meaningful milestone than a 1,000-qubit device that cannot correct errors reliably.
How Zuchongzhi 3.0 compares with Google Willow and IBM Heron
A direct comparison requires care because the three teams use different benchmarks and different error correction codes. Google Willow (72 qubits, announced December 2024) demonstrated below-threshold performance using a surface code and published the result in peer-reviewed form. IBM Heron (133 qubits, 2024) focuses on gate fidelity and device connectivity but has not yet published equivalent below-threshold error correction demonstrations at scale.
Zuchongzhi 3.0 sits broadly in the same experimental tier as Google Willow: both have demonstrated the qualitative milestone of below-threshold error correction on superconducting hardware, with similar qubit counts and competitive fidelity numbers. Neither is anywhere near a fault-tolerant machine; both represent genuine progress toward one. The honest assessment is that China’s superconducting programme is at or near the global frontier — not behind it, but not clearly ahead either.
Beyond Superconducting: China’s Multi-Platform Strategy
One of the most consequential features of China’s quantum programme — examined in detail by the Center for Strategic and International Studies in its January 2026 report Understanding China’s Quest for Quantum Advancement — is its deliberate multi-platform strategy. Rather than concentrating on a single hardware modality, Chinese research institutions and state-funded companies are pursuing superconducting, photonic, trapped-ion, neutral-atom, and silicon spin qubit platforms in parallel. The rationale is hedging: it is not yet clear which architecture will prove most scalable for fault-tolerant computation, and pursuing several reduces the risk of backing the wrong one.
Jiuzhang and photonic quantum computing: 3,050 photons and the limits of boson sampling
Photonic quantum computing uses individual photons — particles of light — as photonic qubits. The Jiuzhang series, also from USTC’s Pan group, is the world’s most prominent photonic quantum computing programme. Jiuzhang 3.0 (results published in 2023) used 255 photons; subsequent iterations have pushed toward 3,050 detected photons in a boson sampling experiment.
Boson sampling is a specific computational task — sampling from the output distribution of indistinguishable photons passing through a linear optical network — that is believed to be classically hard to simulate at large scale. It has no direct practical application but serves as a quantum supremacy benchmark. The Jiuzhang team has consistently argued that its results demonstrate computational advantage over classical supercomputers for this specific task.
The honest caveat, acknowledged even within the quantum computing community, is that boson sampling results are difficult to verify and their classical hardness assumptions have been progressively challenged. In 2022, classical algorithms improved significantly, narrowing the claimed advantage. The 3,050-photon scale makes direct classical simulation harder again, but the claim should be read as “advantage on a narrow, application-free benchmark” rather than “general quantum computational supremacy.” The Jiuzhang line is scientifically impressive; it is not a path to near-term commercial utility or cryptographic relevance.
Trapped ions: Tsinghua’s 512-ion record and its practical significance
Trapped-ion qubits use electrically charged atoms (ions) suspended in electromagnetic fields as qubits. They offer some of the highest fidelity operations available in any hardware platform, along with naturally long qubit coherence times — the duration over which a qubit maintains its quantum state before decoherence destroys the information. The trade-off is slower gate speeds and significant engineering challenges in scaling to many ions.
Tsinghua University’s trapped-ion group reported a 512-ion chain in 2024–2025, claiming a world record for the number of ions simultaneously trapped and manipulated in a single device. This is a scaling milestone for the platform: demonstrating that large ion chains can be controlled without catastrophic crosstalk is non-trivial. However, the number of ions trapped is not the same as the number of high-fidelity qubits available for computation; reports of the actual two-qubit gate fidelity across the full 512-ion chain are more important than the headline count and, as of mid-2025, those figures from peer-reviewed publication remain partial.
The practical significance is that China is investing in trapped-ion systems at a scale that few Western commercial or academic programmes match — with the exception of IonQ and Quantinuum in the United States and United Kingdom respectively. Whether this investment produces near-term computational advantage over superconducting systems is uncertain; the more credible near-term role for trapped-ion systems is in quantum simulation of molecular and materials systems, where high fidelity matters more than raw qubit count.
Neutral atoms: Hanyuan-1, 100-qubit commercial deployment, and the 2,024-atom array
Neutral-atom quantum computing uses uncharged atoms — typically rubidium or caesium — trapped in arrays of laser beams called optical tweezers, as neutral-atom qubits. The platform has attracted significant attention since 2022 because of its combination of long coherence times, native high connectivity between qubits, and apparent scalability to large arrays.
Hanyuan-1, developed by the Beijing-based company Boya Quantum, was announced in 2025 as China’s first commercially deployed neutral-atom quantum computer, offering 100 qubits via cloud access. Its significance is twofold: it is a commercial deployment rather than a laboratory demonstration, and it places China in a competitive position relative to US neutral-atom startups such as QuEra and Atom Computing, which have been building toward similar milestones.
Separately, researchers at Peking University reported a 2,024-atom array in 2024–2025 — a number chosen symbolically to echo the year — demonstrating that neutral-atom platforms can be physically scaled to very large numbers of sites. The gap between 2,024 atoms in an array and 2,024 high-fidelity qubits available for computation is substantial: the fidelity of entangling gates across the full array was not uniformly high, and most useful computation on neutral-atom hardware currently uses a subset of available sites. Still, the scaling demonstration is meaningful for the platform’s long-term trajectory.
Silicon spin qubits: China’s world-first claim assessed
Silicon spin qubits encode quantum information in the spin state of individual electrons or nuclei in silicon — the same material used in classical semiconductor manufacturing. The appeal is compatibility with existing fabrication infrastructure and potentially very small physical footprint. The challenge is that spin qubits are extremely sensitive to materials imperfections and require precise control of individual dopant atoms.
Chinese research groups, including teams at Peking University, have claimed world-first results in specific silicon spin qubit configurations in 2024–2025. Independent assessment of these claims requires peer-reviewed publication with full device characterisation; some results have been published in preprint form. The field of silicon spin qubits globally remains behind superconducting and trapped-ion systems in overall device performance, and China’s position within it is competitive but not dominant. This is the platform where claims require the most caution: “world-first” in a narrowly defined configuration can be technically accurate while being strategically less significant than it sounds.
Having established the hardware picture across all five platforms, the next question is what the surrounding commercial and investment ecosystem looks like — and whether state funding is translating into deployable capability.
The Commercial Ecosystem: From Lab to Market
Hardware milestones do not automatically translate into commercial or strategic capability. The path from a laboratory processor to a deployable system requires software stacks, error mitigation techniques, application development, and — for security-critical use cases — regulatory and integration frameworks. China’s quantum ecosystem has expanded significantly in all these dimensions since 2020, but unevenly.
Origin Quantum and Tianyan-504: the state-backed commercial push
Origin Quantum, founded in 2017 and headquartered in Hefei — China’s self-styled “quantum city” — is the country’s leading quantum computing company by public profile and state backing. Its Tianyan-504 superconducting processor (announced 2024–2025) represents a 504-qubit system positioned as a cloud-accessible commercial platform. Origin Quantum offers access through its Quafu cloud service and has partnerships with Chinese state enterprises and universities.
The Tianyan-504 qubit count is notable but, as emphasised throughout this article, qubit count must be read alongside fidelity and error correction data. Published benchmarks for Tianyan-504 show two-qubit gate fidelities in the 99.5% range — competitive but not at the frontier established by Zuchongzhi 3.0 or Google Willow. The commercial value of a 504-qubit system with current error rates lies primarily in near-term noisy intermediate-scale quantum (NISQ) applications: quantum simulation for chemistry and materials, certain optimisation problems, and quantum machine learning experiments, all of which are active research areas rather than proven commercial use cases.
Quantum cloud access is the key commercial metric here: Chinese researchers and enterprises now have domestic access to quantum hardware without routing through US providers such as IBM Quantum or AWS Braket — a meaningful change for both research velocity and supply chain independence.
China’s $15 billion+ public investment: what it is — and is not — buying
China’s cumulative public investment in quantum technologies — across quantum computing, quantum communications, and quantum sensing — is estimated by various sources including CSIS (January 2026) at over $15 billion since the launch of major national programmes in 2016, with quantum computing receiving a substantial share. The 14th Five-Year Plan (2021–2025) and subsequent planning documents have maintained quantum technology as a national strategic priority.
What this investment is buying: research infrastructure (the National Laboratory for Quantum Information Sciences in Hefei is one of the largest quantum research facilities in the world), a pipeline of trained researchers, multiple hardware platforms at competitive fidelity levels, and a domestic commercial ecosystem that did not exist a decade ago.
What it is not straightforwardly buying: guaranteed technological leadership. Investment in quantum computing does not map linearly to outcomes because the key bottlenecks — qubit coherence, error correction at scale, software and algorithm development — are scientific and engineering problems that money accelerates but does not solve on a fixed schedule. The CSIS January 2026 report notes that China’s innovation ecosystem shows regional concentration (Hefei, Beijing, Shanghai) and strong academic output but relative weakness in the translation from research to deployable systems, a gap that the commercial sector is only beginning to address.
Comparison with US investment is complicated by classification and accounting methodology. The US National Quantum Initiative Act (2018) authorised roughly $1.2 billion over five years in federal funding, but total US quantum investment — including DARPA programmes, NSF grants, DOE national laboratory spending, and private venture capital — significantly exceeds that figure. On a government-only basis, China likely outspends the United States; on a total ecosystem basis including private capital, the gap is less clear and possibly inverted.
Which country leads in quantum computing right now? A calibrated answer
This is the most frequently asked question about the field, and the honest answer is that “leads” requires disaggregation by dimension.
On superconducting hardware quality at the research frontier, the United States (Google, IBM) and China (USTC/Zuchongzhi) are in the same experimental tier. Neither has a decisive advantage; each has achieved below-threshold error correction on small systems.
On quantum communications and quantum key distribution (QKD), China holds a clear lead. The Micius satellite-based QKD network and the Beijing–Shanghai quantum communication backbone are the world’s most extensive deployed quantum communication infrastructure. This is a separate field from quantum computing but contributes to China’s overall quantum technology posture.
On commercial quantum computing ecosystem — software tools, developer community, application cases, integration with classical computing infrastructure — the United States leads significantly. IBM Quantum, Google Quantum AI, and the broader US startup ecosystem (IonQ, Quantinuum, QuEra, PsiQuantum, and others) have more mature software stacks and larger user communities.
On quantum investment breadth across hardware modalities, China’s multi-platform strategy means it has competitive programmes in more areas simultaneously than any single US company, though the US national laboratory and university ecosystem provides equivalent breadth at the system level.
The summary assessment: no single country leads unambiguously across all dimensions. China leads in QKD and communications; the United States leads in commercial software and ecosystem maturity; both are competitive at the hardware research frontier for computing. This calibrated picture is less satisfying than a simple ranking but more accurate.
The Geopolitical and Security Dimension
Having established the hardware picture, the next question — the one that most drives policy concern — is what China’s quantum progress means for global security. This section addresses that question carefully, distinguishing between what is demonstrably true today, what the trajectory suggests, and what remains genuinely uncertain.
The CRQC threat to current encryption: what the USCC assessment actually says
The US–China Economic and Security Review Commission’s November 2025 report Vying for Quantum Supremacy is the most recent authoritative US government assessment of the CRQC threat. A cryptographically relevant quantum computer is defined as a device capable of executing Shor’s algorithm at a scale sufficient to factor the large prime numbers underpinning RSA and elliptic-curve cryptography — the encryption standards protecting most internet traffic, financial transactions, and government communications today.
The USCC assessment is explicit that no CRQC currently exists and that building one would require millions of physical qubits with error rates significantly below current demonstrated levels. The report does, however, flag two concerns. First, the trajectory of capability improvement — particularly below-threshold error correction demonstrations — means the timeline to a CRQC is shortening, even if it remains measured in years to decades rather than months. Second, the practice of harvest now, decrypt later — in which adversaries collect encrypted data today to decrypt once a CRQC becomes available — creates a present-day threat even without an existing CRQC. Data with long-term sensitivity (intelligence assessments, health records, weapons specifications) is at risk from this strategy now.
The Q-Day concept — the notional date on which a CRQC first breaks operational encryption — is genuinely uncertain. Estimates from credible institutions range from the early 2030s to beyond 2040, with significant expert disagreement. Presenting any single date as authoritative would misrepresent the state of knowledge.
Post-quantum cryptography: NIST standards and the race to deploy them
The response to the long-term CRQC threat is post-quantum cryptography (PQC) — cryptographic algorithms based on mathematical problems believed to be hard for quantum computers as well as classical ones. The US National Institute of Standards and Technology finalised its first PQC standards in August 2024 (FIPS 203, 204, and 205), based on algorithms including CRYSTALS-Kyber for key encapsulation and CRYSTALS-Dilithium and SPHINCS+ for digital signatures.
These NIST PQC standards represent the primary recommended migration path for most organisations. The challenge is deployment at scale. Cloudflare’s October 2025 report State of the Post-Quantum Internet provides rare real-world deployment data: post-quantum TLS (Transport Layer Security) adoption has accelerated significantly since 2024, with major browsers and a growing share of web servers now supporting hybrid classical/post-quantum handshakes. However, migration of internal systems, legacy infrastructure, and non-web protocols lags substantially behind the public web.
Quantum-safe encryption and quantum-safe algorithms are terms used interchangeably with PQC in most policy contexts. Quantum key distribution (QKD) — a separate, hardware-based approach using quantum physics to distribute cryptographic keys — is sometimes proposed as an alternative, but most cryptographic authorities including NIST and the UK’s NCSC treat QKD as a complement rather than a replacement for PQC, given its infrastructure requirements and deployment constraints.
US export controls, the Entity List, and their real impact on China’s progress
US export controls, including Entity List designations targeting Chinese quantum computing firms and expanded controls on quantum computing-related equipment and software from 2023 onward, represent a deliberate effort to slow China’s quantum progress by restricting access to advanced fabrication tools, cryogenic equipment, and some categories of software.
The real impact is mixed. For superconducting quantum computing specifically, the most critical fabrication equipment — dilution refrigerators, electron-beam lithography systems — has multiple suppliers including European ones (Oxford Instruments, Bluefors) that are not subject to US controls. China has also invested heavily in domestic alternatives; Zuchongzhi and Tianyan-504 were fabricated using domestic clean-room facilities, suggesting that the most restrictive controls have not yet created a decisive bottleneck in hardware fabrication.
Where controls may be more effective is in quantum computing software, classical control electronics, and advanced semiconductor components used in qubit control systems. These are less visible areas than the qubit count but are genuine bottlenecks. The CSIS January 2026 report identifies classical control electronics as one of the underappreciated constraints on China’s quantum scaling roadmap.
Quantum espionage — state-sponsored theft of quantum computing research — is noted in multiple US government assessments as a vector through which controls can be partially circumvented. This is a persistent concern across all advanced technology competition, not unique to quantum computing.
US–China quantum competition: where each side holds genuine advantages
The US advantages in quantum computing competition are: a larger and more mature commercial ecosystem; deeper integration between academic research and commercial development; a broader talent pipeline drawing from global university recruitment; stronger software and algorithm development; and an established lead in error correction theory, where most foundational work originates from US and European academic groups.
China’s advantages are: larger and more sustained government funding with clearer long-term strategic commitment; a multi-platform hardware programme with no equivalent in any single US institution; a domestic QKD deployment that gives its defence and intelligence services quantum-secured communications at a scale no Western country has matched; and a research community of comparable size and quality in hardware-focused quantum science.
Neither side has a decisive advantage that makes the outcome of competition predictable. The more accurate framing, consistent with the USCC November 2025 assessment, is that the two countries are running in parallel on most hardware dimensions, with the United States ahead on commercialisation and China ahead on state-directed deployment of quantum communications infrastructure.
What China Still Lacks — and the Honest Assessment of the Gap
This section addresses what is perhaps the most important question for accurate assessment: given the genuine milestones above, how far is China — and the global field — from quantum computing that matters strategically?
The distance between 105 physical qubits and a fault-tolerant machine
Fault-tolerant quantum computing requires logical qubits — error-protected units — rather than raw physical qubits. The physical-to-logical qubit ratio is the central scaling challenge: current leading error correction schemes (surface codes) require roughly 1,000–10,000 physical qubits per logical qubit, depending on the target error rate and the physical error rate of the underlying hardware.
This means that running Shor’s algorithm to break 2048-bit RSA — estimated to require approximately 4,000 logical qubits — would need on the order of four million to forty million physical qubits at current error rates. Zuchongzhi 3.0 has 105 physical qubits demonstrating below-threshold behaviour on a small number of logical qubits. The gap between 105 and four million is not merely quantitative; it involves engineering challenges in wiring, cooling, control electronics, and fabrication uniformity that no research group has yet solved at any scale.
The logical qubit overhead is not a fixed number: it improves as physical error rates improve and as error correction codes become more efficient. The progress demonstrated by Google Willow and Zuchongzhi 3.0 in below-threshold operation does reduce the overhead required, but the reduction is incremental rather than transformative at this stage. A realistic engineering path to millions of physical qubits with the required connectivity and uniformity does not currently exist in any publicly disclosed roadmap, including those of IBM and Google.
Software, algorithms, and talent: the less-discussed bottlenecks
The quantum software stack — compilers, simulators, error mitigation libraries, application frameworks — is a significant constraint that receives far less attention than hardware. Writing efficient quantum algorithms for near-term noisy hardware is a specialised skill; adapting classical business problems into quantum-amenable formulations is harder still. China’s quantum software ecosystem, while growing, lags behind the US ecosystem in tooling maturity, open-source community size, and integration with classical cloud infrastructure.
Talent concentration is a related bottleneck. While China produces a large number of physics and engineering PhDs with quantum-relevant training, the top tier of quantum error correction theorists and algorithm developers is still disproportionately concentrated in US and European institutions. China retains some of this talent domestically and has been effective at recruiting overseas Chinese researchers, but the global distribution of expertise at the very frontier of quantum algorithm and error correction research remains weighted toward Western institutions.
What ‘quantum supremacy’ claims do and do not prove
The distinction between quantum supremacy and quantum advantage matters here. Quantum supremacy — performing any computation beyond classical reach, however narrowly defined — has been demonstrated credibly by both Google (2019, random circuit sampling) and the Jiuzhang team (boson sampling). These results are scientifically meaningful: they demonstrate that quantum hardware can, in specific and engineered conditions, outperform classical simulation.
What they do not prove: that the computation performed has any practical value; that the same hardware can be repurposed for commercially or strategically relevant tasks; that the quantum advantage demonstrated scales to larger, more complex problems of practical interest. Quantum supremacy on a purpose-built benchmark is analogous to a prototype aircraft flying in ideal conditions on a test track — impressive engineering, but a long way from operational deployment.
The more meaningful framing for assessing China’s position is quantum advantage in useful tasks — and here, no system globally, Chinese or otherwise, has yet demonstrated convincing quantum advantage on a problem of genuine commercial or strategic relevance. The field is working toward this; China is among the leaders working toward it; it has not been achieved.
What These Milestones Mean for Businesses and Policymakers
The analytical picture established above has concrete implications for organisations making decisions now. This section converts the technical assessment into actionable framing — without overstating the urgency or understating the real preparation required.
Timelines: realistic windows for quantum advantage in useful tasks
The quantum advantage timeline for commercially relevant applications — drug discovery, materials simulation, financial optimisation, logistics — is the subject of genuine expert disagreement. Conservative estimates place meaningful quantum advantage in these domains in the early-to-mid 2030s at the soonest, contingent on continued progress in error correction scaling. More optimistic estimates from hardware companies suggest late 2020s for narrow simulation tasks. Both should be treated as estimates rather than forecasts; the history of the field is one of consistent underestimation of near-term engineering difficulty alongside genuine long-run progress.
For the specific case of cryptographic relevance — a CRQC that threatens RSA and elliptic-curve encryption — most credible estimates place the risk beyond 2030 and possibly beyond 2035. The USCC November 2025 report does not commit to a specific date but treats the 2030s as the planning horizon for which governments and critical infrastructure operators should be completing cryptographic migration.
A quantum computing roadmap for any organisation should therefore plan on two separate timelines: a longer horizon (2030–2040) for transformative computational applications, and a nearer horizon (now–2030) for cryptographic migration driven by the harvest-now-decrypt-later threat.
What organisations should actually do now about post-quantum security
The most pressing action — and the one that does not require waiting for quantum computers to arrive — is cryptographic migration. Organisations should:
- Conduct a quantum risk assessment: inventory all systems using public-key cryptography (RSA, ECC, DH) and identify data assets with long-term sensitivity that could be harvested today.
- Begin transitioning public-key infrastructure to NIST-standardised PQC algorithms (FIPS 203/204/205), starting with internet-facing services where hybrid deployments are already supported by major TLS libraries and browsers.
- Engage with supply chain and vendor dependencies: many organisations