Skip to content
Home » Articles » Quantum Machine Learning and Data Assurance: Securing the Black Box

Quantum Machine Learning and Data Assurance: Securing the Black Box

Understanding Data Assurance and Integrity

In the face of the extraordinary expansion of artificial intelligence, the necessity to safeguard data integrity and comprehend its provenance has taken on an unprecedented level of importance. The emergence of quantum computing and quantum machine learning has added a new layer to this issue. Quantum machine learning, a convergence of quantum computing and machine learning, is set to revolutionise data processing and analysis. However, these advancements necessitate robust data assurance mechanisms, especially when dealing with large datasets of images, videos and signals that may have been compromised [1,2].

The Importance of Data Provenance and Integrity

Data provenance refers to the origin of data, its movement, and how it has been used and modified over time. Understanding data provenance is crucial in ensuring the integrity of data, especially when dealing with complex datasets used in machine learning. It provides insights into how these datasets have been created and used to train classical AI and quantum machine learning models. This understanding is particularly important when the datasets may have been compromised.

Data integrity ensures that data is accurate, consistent, and reliable throughout its lifecycle. In the context of quantum machine learning, ensuring data integrity is paramount for the reliability of the models.

Quantum Machine Learning and Data Assurance

Quantum machine learning leverages quantum mechanics to revolutionise machine learning algorithms [3], utilising quantum bits or qubits that can exist in multiple states simultaneously. This property, known as superposition, enables quantum computers to process vast amounts of data at unprecedented speeds, potentially solving complex problems currently unmanageable for classical computers. Quantum machine learning can expedite tasks such as pattern recognition, anomaly detection, and optimisation in large datasets, making it a game-changer for many sectors.

However, ensuring the integrity and security of the data used in quantum machine learning is crucial. Unlike classical machine learning, which processes data directly and follows set rules, quantum machine learning requires transforming data into a special quantum state to analyse it [4]. This complexity demands robust data assurance practices to guarantee the quality and integrity of the data used in machine learning models. Data assurance is particularly important in sectors where data security and privacy are paramount, such as defence, critical infrastructure, finance, and healthcare. As quantum computing and applications continue to evolve, data assurance will play a critical role in ensuring compliance with regulatory standards and protecting sensitive information.

Trust Levels for both Classical and Quantum Data

Trust levels provide a framework for organisations to assess and improve their data assurance practices as they adopt quantum machine learning technologies [5]. However, it’s important to note that the specific measures and requirements at each level may vary depending on the specific use case and regulatory environment. As organisations progress through these levels, they can ensure that their data assurance practices keep pace with their quantum machine learning capabilities. This will help them to maintain the integrity and security of their data as they leverage the power of quantum machine learning.

Level 5 – Complete Trust – ensures robust data assurance mechanisms are in place for both classical AI and quantum data. This includes comprehensive data integrity checks, full data provenance tracking, advanced quantum-specific data assurance measures, and robust data confidentiality and availability measures. At this level, the data assurance measures are expected to comply with regulatory standards and are regularly audited for compliance.


Quantum machine learning holds immense promise for the future. However, to fully realise this potential, robust data assurance mechanisms are essential. As we continue to explore the quantum domain, let’s also invest in ensuring the integrity and understanding the provenance of our quantum data. After all, in the era of quantum machine learning, data assurance is not just a good practice – it’s a necessity. This will ensure that as we step into the future of quantum machine learning, we do so with confidence and security. It is particularly important when the datasets may have been compromised by adversarial attacks, misadventure, or erroneously entered data at the creation of the training set for classical AI systems [1,2]. By addressing these challenges head-on, we can pave the way for a more secure and reliable quantum future.

About Gary Morgan: Gary Morgan is an experienced board director, chief executive, consultant, and corporate advisor with extensive experience in strategy, innovation, and growth across various deep tech sectors including health tech, agtech, information security, and research. He is a Fellow at the Governance Institute of Australia and serves on the Griffith University Industry Advisory Board for the ICT School. Gary has co-authored papers and reports published in entrepreneurship and medical journals.

Acknowledgment: I would like to thank Associate Professor Muhammad UsmanDr Akib Karim and Dr Peter Kambouris for their invaluable input and feedback. This article was crafted with the assistance of AI technology.


1.West, M.T., Tsang, SL., Low, J.S., Hill, C.D., Leckie, C., Hollenburg, L.C., Erfani, S.M., & Usman, M. Towards quantum enhanced adversarial robustness in machine learning. Nat Mach Intell 5, 581–589 (2023).

2.Trend, A., Usman, M., Can quantum computing protect AI from cyber attacks? CSIRO (2023).

3.Editorial. Seeking a quantum advantage for machine learning. Nat Mach Intell 5, 813 (2023).

4. Huang, HY., Broughton, M., Mohseni, M. et al. Power of data in quantum machine learning. Nat Commun 12, 2631 (2021).

5.Hamada, K. et al, Guidelines for quality assurance of machine learning-based artificial intelligence (2020).