Recent quantum circuit obfuscation techniques supported by trusted hardware aim to protect algorithms from untrusted cloud providers, yet structural patterns often remain observable. In this work, we leverage Machine Learning (ML) to learn algorithmic footprints from unobfuscated circuits and apply them to analyze obfuscated circuits, revealing residual structural signatures. Our ML framework not only identifies the quantum algorithm class but also strips the obfuscated circuits from the additional gates used for obfuscation to recover the original quantum circuit. Our approach provides a flexible framework to evaluate the effectiveness of diverse obfuscation methods and quantify how much algorithmic information persists despite structural transformations.
Recent quantum circuit obfuscation techniques supported by trusted hardware aim to protect algorithms from untrusted cloud providers, yet structural patterns often remain observable. In this work, we leverage Machine Learning (ML) to learn algorithmic footprints from unobfuscated circuits and apply them to analyze obfuscated circuits, revealing residual structural signatures. Our ML framework not only identifies the quantum algorithm class but also strips the obfuscated circuits from the additional gates used for obfuscation to recover the original quantum circuit. Our approach provides a flexible framework to evaluate the effectiveness of diverse obfuscation methods and quantify how much algorithmic information persists despite structural transformations.
The developments in quantum computing hardware and software are focused on boosting the scalability and robustness of quantum systems. With near-term quantum computers poised to accelerate specific computational tasks, ensuring their security has become a pressing priority. Recent efforts have proposed practical solutions to ensure data confidentiality in quantum computers accessed through the cloud, particularly in the presence of untrusted quantum cloud providers by leveraging trusted hardware that enables quantum circuit obfuscation. However, these solutions overlook the detailed structural information of the quantum algorithms, which may limit the effectiveness of the obfuscation methods. In this paper, we investigate the structural constraints of quantum algorithms executed across different quantum architectures. Based on this analysis, we develop a Machine Learning (ML)-based classifier that reveals the underlying quantum algorithm of the obfuscated physical quantum circuit. We also propose a pre-processing phase to minimize the obfuscation's impact on the classifier accuracy. We evaluate our proposed ML-based classification model on different obfuscated physical quantum circuits, which are transpiled to support IBM quantum hardware. Our analysis shows the vulnerability of existing obfuscation methods to reverse engineering attacks, which can potentially expose sensitive information about quantum circuits.
Circuit cutting is a novel technique in the field of quantum computing that offers a promising approach for breaking up large quantum circuits into smaller ones that can be executed on smaller quantum computers. Circuit cutting offers a scalable and hybrid computing approach, by combining classical and quantum computers to run large quantum circuits. At the same time, quantum circuits represent the fundamental building blocks of quantum algorithms, and obfuscation techniques are crucial for protecting the sensitive information and intellectual property encoded in these circuits. This paper constitutes the first security analysis of circuit cutting techniques as a means to protect quantum circuits, and not just for cutting bigger quantum circuits to fit on smaller devices. As cloud providers are in control of quantum computers and give access to users, utilizing cloud-based quantum computers for executing quantum circuits involves considerable security risks. This work demonstrates that quantum circuits can be effectively obfuscated by using quantum circuit cutting techniques. If users are worried about an adversary who attempts to reverse-engineer the original quantum algorithm, circuit cutting can be used as a security measure. To further provide configurable obfuscation level, i.e. number of possible circuits that the cloud provider would have to guess, this work proposes a novel dummy-subcircuits technique. This technique introduces dummy subcircuits with additional qubit cut points to create more confusion for the cloud provider and further help to obfuscate the user's original circuits with limited cost.
We introduce ObfusQate, a novel framework that performs software obfuscation using quantum primitives to enhance the security of both classical and quantum programs. ObfusQate implements eight obfuscation techniques categorized into quantum circuit-level and code-level transformations. Circuit-level obfuscation leverages quantum gates, identity operations, and unitary transformations to construct non-intuitive circuits that resist reverse engineering. Code-level obfuscation employs quantum-based opaque predicates to obscure control flow, making execution paths unpredictable. We evaluate the effectiveness of ObfusQate using three benchmark quantum algorithms: Bernstein-Vazirani, Shor’s, and QAOA. Experiments show that the framework incurs tunable polynomial overhead in both time and space, while preserving functional correctness. Additionally, we demonstrate an offensive use case by embedding a keylogger into Shor’s algorithm. Large language models (GPT-4o, GPT-o3-mini, and Grok-3) successfully identified the payload in the unobfuscated version, but failed when it was obfuscated using ObfusQate. To our knowledge, ObfusQate is the first automated system for quantum software obfuscation. It provides a foundation for secure deployment of quantum software in adversarial and shared environments.
Machine Learning (ML) has been widely applied across numerous domains due to its ability to automatically identify informative patterns from data for various tasks. The availability of large-scale data and advanced computational power enables the development of sophisticated models and training strategies, leading to state-of-the-art performance, but it also introduces substantial challenges. Quantum Computing (QC), which exploits quantum mechanisms for computation, has attracted growing attention and significant global investment as it may address these challenges. Consequently, Quantum Machine Learning (QML), the integration of these two fields, has received increasing interest, with a notable rise in related studies in recent years. We are motivated to review these existing contributions regarding quantum circuit-based learning models for classical data analysis and highlight the identified potentials and challenges of this technique. Specifically, we focus not only on QML models, both kernel-based and neural network-based, but also on recent explorations of their integration with classical machine learning layers within hybrid frameworks. Moreover, we examine both theoretical analysis and empirical findings to better understand their capabilities, and we also discuss the efforts on noise-resilient and hardware-efficient QML that could enhance its practicality under current hardware limitations. In addition, we cover several emerging paradigms for advanced quantum circuit design and highlight the adaptability of QML across representative application domains. This study aims to provide an overview of the contributions made to bridge quantum computing and machine learning, offering insights and guidance to support its future development and pave the way for broader adoption in the coming years.
The success of quantum circuits in providing reliable outcomes for a given problem depends on the gate count and depth in near-term noisy quantum computers. A circuit (that implements a given function) with a low gate count and short depth is more likely to give a correct solution than the circuit variant with a higher gate count and depth. As such, quantum circuit compilers that decompose high-level gates to native gates of the hardware and optimize the circuit play a key role in quantum computing. However, the quality and time complexity of the optimization process can vary significantly especially for practically relevant large-scale quantum circuits. As a result, third-party (often less-trusted/untrusted/unreliable) compilers have emerged, claiming to provide better and faster optimization of complex quantum circuits than so-called trusted compilers. However, untrusted compilers, with their varying adversarial motives ranging from semi-honest to malicious can pose severe security risks such as counterfeiting the sensitive intellectual property (IP) embedded within the quantum circuit. We propose an obfuscation technique for quantum circuits using randomized reversible gates to protect them from such attacks during compilation. The idea is to insert a small random circuit into the original circuit and send it to the untrusted compiler. Since the circuit function is corrupted, the adversary (i.e., an untrusted compiler) may get incorrect IP. However, the user may also get incorrect output post-compilation (even though highly optimized). To circumvent this issue, we concatenate the inverse of the random circuit in the compiled circuit to recover the original functionality. We demonstrate the practicality of our method by conducting exhaustive experiments on a set of benchmark circuits and measuring the quality of obfuscation by calculating the Total Variation Distance (TVD) and Degree of Functional Corruption (DFC) metrics. Our method achieves TVD of up to 1.92, indicating substantial functional corruption, and performs at least2xbetter than a previously reported obfuscation method. We also attained a DFC as low as −0.75, approaching the −1 score that represents maximal functional corruption, further ensuring the robustness of our proposed method against counterfeiting. We also propose a novel adversarial reverse engineering (RE) approach and show that the proposed obfuscation is resilient against RE attacks. The proposed technique introduces minimal degradation in fidelity (∼1%to∼3%on average).
Quantum circuit obfuscation is becoming increasingly important to prevent theft and reverse engineering of quantum algorithms. As quantum computing advances, the need to protect the intellectual property contained in quantum circuits continues to grow. Existing methods often provide limited defense against structural and statistical analysis or introduce considerable overhead. In this paper, we propose a novel quantum obfuscation method that uses randomized U3 transformations to conceal circuit structure while preserving functionality. We implement and assess our approach on QASM circuits using Qiskit AER, achieving over 93\% semantic accuracy with minimal runtime overhead. The method demonstrates strong resistance to reverse engineering and structural inference, making it a practical and effective approach for quantum software protection.