Quantum Machine Learning Assesses Post-Quantum Protocol Resilience

Institute of Theoretical and Applied Informatics, Polish Academy of Sciences is applying quantum machine learning in a new way: not creating new cryptography, but proactively testing the defenses of existing protocols. The work details the successful demonstration of loading the probability distribution of hash-based digital signatures into quantum computer memory using Quantum Generative Adversarial Networks, or QGANs. This represents a concrete step toward utilizing quantum computing to actively attack these signatures, even with the limitations of near-term hybrid quantum-classical methods. The team views this approach as a first step in the workflow for utilizing quantum computing to attack post-quantum cryptographic primitives, moving beyond simply designing algorithms resistant to known quantum threats like Shor’s and Grover’s algorithms.

Post-Quantum Cryptography & Quantum Threat Landscape

Quantum Generative Adversarial Networks (QGANs) are now being directed toward proactively testing the defenses of post-quantum cryptography, rather than solely focusing on creating new cryptographic methods. This shift signifies a growing recognition that current quantum computers can reveal vulnerabilities in algorithms designed to withstand future, more powerful machines. Researchers are no longer solely concerned with building unbreakable codes; they are actively probing for weaknesses in those already proposed as quantum-resistant. This achievement isn’t about breaking the signatures now, but about establishing the capability to do so, and understanding how near-term quantum devices can be leveraged for attack. The work confirms that “near-term hybrid quantum-classical methods possess capabilities required for this purpose,” indicating that even current technology can be used to analyze and potentially compromise post-quantum schemes.

This is a departure from simply theorizing about the future threat posed by cryptographically relevant quantum computers (CRQCs), which require a large number of qubits, long coherence times, and high-fidelity quantum gates. The researchers present an example application of QGANs for hash-based digital signatures, utilizing them to model their underlying probability distributions. This allows a quantum computer to essentially “learn” the signature’s characteristics, potentially revealing patterns or weaknesses that classical analysis might miss. The team’s methodology builds upon the foundation of classical Generative Adversarial Networks (GANs), replacing components with quantum neural networks to leverage quantum phenomena like superposition and entanglement. Ultimately, the goal is to ensure the long-term security of these protocols by identifying and addressing vulnerabilities before they can be exploited, and this work provides a crucial initial framework for doing so.

Shor & Grover Algorithms Impact on Public-Key Systems

The field of cryptographic security is undergoing a fundamental shift, driven not by the immediate cracking of current encryption, but by proactive assessment of vulnerabilities using emerging quantum tools. While fully fault-tolerant quantum computers capable of breaking widely used public-key systems remain years away, researchers are now actively employing near-term quantum devices to scrutinize the resilience of post-quantum cryptography (PQC) protocols. Recent work demonstrates a surprising application of quantum machine learning, specifically Quantum Generative Adversarial Networks (QGANs), not to create new cryptography, but to rigorously test existing PQC implementations. Researchers present an example application of QGANs for loading the probability distribution of hash-based digital signatures into the memory of a quantum computer, a concrete step towards utilizing quantum computing to attack these signatures. This isn’t about breaking signatures now, but about establishing a workflow for future analysis.

The core of this approach lies in leveraging the unique capabilities of QGANs. Classical GANs involve two neural networks, a generator and a discriminator, trained in competition. QGANs replace these with quantum neural networks, offering the potential to efficiently encode complex probability distributions. This is a significant departure from earlier investigations. The researchers argue this initial step is crucial for ensuring the long-term security of these protocols as quantum computing technology matures and the threat landscape evolves. The rapid adoption of PQC standards, evidenced by inclusion in Java 26, released in March 2026, and OpenSSH 10.0 suite, underscores the urgency of this research.

Jaroslaw A. Miszczak from the Institute of Theoretical and Applied Informatics, Polish Academy of Sciences is investigating a new direction in quantum security research; rather than focusing on building quantum computers powerful enough to break existing encryption, his work investigates how even near-term machines can be used to test the resilience of proposed post-quantum cryptographic protocols. This represents a shift from purely defensive strategies, designing algorithms resistant to future quantum attacks, to a proactive, offensive approach focused on identifying vulnerabilities in current standards. This isn’t merely theoretical exploration; the researchers are actively working to model the behavior of cryptographic primitives on quantum hardware. This is particularly significant given the ongoing development and standardization of post-quantum cryptographic algorithms, including recent additions to OpenSSH 10.0 suite and Java, which are becoming increasingly prevalent in cloud environments and general-purpose programming languages. Java 26, released in March 2026, introduces post-quantum-ready JAR signing and hybrid public key encryption support to prepare applications for the quantum era. The distinction between NISQ computers and their future counterparts, Cryptographically Relevant Quantum Computers (CRQCs) is crucial.

The field of cryptographic security is witnessing a shift; attention is now focused on proactively testing the resilience of post-quantum protocols with the limited quantum computers available rather than solely concentrating on developing entirely new cryptographic systems. The core of this strategy involves utilizing QGANs to model and analyze the probability distributions inherent in post-quantum cryptographic primitives. QGANs translate this concept to the quantum realm, employing parameterized quantum circuits to potentially enhance the generative capabilities and efficiency of the process. This research isn’t about immediately breaking post-quantum signatures, but rather establishing a first step in the workflow enabling the utilization of quantum computing for attacking post-quantum cryptographic primitives. By employing QGANs, they can effectively “learn” the characteristics of a signature scheme, allowing for targeted attacks and the identification of potential vulnerabilities before they can be exploited by a more powerful quantum computer. This example application with hash-based signatures suggests that QGANs could be adapted to analyze other post-quantum cryptographic schemes, providing a valuable tool for strengthening the security of future communication systems.

The assumption that quantum computing’s initial impact on cryptography will be about breaking codes is proving increasingly narrow; the focus is shifting toward proactively assessing the vulnerabilities of proposed post-quantum solutions, even with the limited quantum hardware available now. The researchers aimed to explore whether quantum machine learning could circumvent these protections. The presented approach can be used as a first step in the workflow enabling the utilization of quantum computing for attacking post-quantum cryptographic primitives. This initial success paves the way for more sophisticated quantum-assisted attacks on post-quantum cryptography, demanding continuous evaluation and refinement of these emerging security standards.