How your favourite PETs can enable secure automation

24 Jun 2024

Image: © chendongshan/

Privacy-enhancing technologies (PETs) are becoming increasingly vital as we navigate the complexities of modern automation and data regulation, argues Zama’s Jeremy Bradley.

Click here for the full Automation Week series.

In an era where data is the lifeblood of innovation, balancing automation and privacy is crucial.

Privacy-enhancing technologies (PETs) are emerging as pivotal tools that empower automation while safeguarding sensitive information.

PETs offer the promise of harnessing data’s full potential without compromising privacy. Let’s look at the symbiotic relationship between PETs and automation, and how they collectively advance a future where data-driven decisions are both effective and secure.

What are privacy-enhancing technologies?

Privacy-enhancing Technologies encompass a range of tools and methodologies designed to protect personal data throughout its lifecycle. These technologies ensure that privacy is maintained during data collection, processing and sharing, making them integral to modern automation solutions.

Here are some of the key PETs:

Differential privacy

This adds controlled noise to datasets, making it difficult to identify individual data points while allowing meaningful analysis.

Homomorphic encryption

This allows computations on encrypted data, enabling secure processing without exposing raw data.

Secure multi-party computation (SMPC)

This enables collaborative computation across multiple parties without revealing their individual inputs.

Federated learning

This is the training of machine learning models across decentralised devices while keeping data localised, thus preserving user privacy.


This removes or modifies personal identifiers from datasets, reducing the risk of re-identification.

Enabling automation with PETS

Automation relies heavily on data-driven decision-making to optimise processes and deliver personalised experiences.

PETs allow organisations to leverage rich datasets while maintaining user privacy. For example, in marketing automation, differential privacy enables the analysis of customer behaviour without exposing individual identities, facilitating targeted campaigns that respect consumer privacy.

In sectors such as healthcare and finance, automated systems require access to sensitive information from multiple sources. PETs such as SMPC and homomorphic encryption enable secure data sharing and collaborative analysis across organisations. This allows automated systems to provide insights and recommendations without compromising the confidentiality of the underlying data. For instance, in automated credit scoring, multiple financial institutions can collaboratively analyse credit data without revealing individual customer details.

Federated learning is revolutionising how automated systems learn and adapt. In scenarios such as autonomous driving, federated learning enables vehicles to collectively improve their models based on data from numerous sources without exchanging raw data. This ensures that automation systems evolve and enhance their capabilities while preserving user privacy.

Analytics is a cornerstone of automation, providing insights that drive decision-making. PETs enable privacy-preserving analytics, allowing organisations to extract value from data without exposing sensitive information. For example, in automated supply chain management, differential privacy can be used to analyse inventory levels and predict demand without revealing specific supplier details.

What are the benefits of integrating PETs with automation?

Incorporating PETs into automated systems enhances trust among users and stakeholders. When users know their data is protected, they are more likely to engage with automated services. This trust is particularly vital in sectors where data sensitivity is paramount, such as healthcare and finance.

Privacy regulations such as the GDPR and CCPA mandate strict data protection measures. PETs enable automated systems to comply with these regulations by ensuring that data privacy is maintained throughout automated processes. This reduces the risk of legal penalties and enhances the reputation of organisations.

PETs allow organisations to innovate by leveraging data in ways that were previously constrained by privacy concerns. This fosters a competitive advantage, enabling businesses to develop new automated solutions and services that differentiate them in the market. For example, privacy-preserving customer analytics can lead to the development of more personalised and effective automated customer service solutions.

Automated systems equipped with PETs can handle large volumes of data securely and efficiently. This scalability is essential for applications like automated fraud detection, where real-time analysis of vast datasets is required to identify anomalies. PETs ensure that such systems can scale without compromising data security.

What are the challenges to consider?

Implementing PETs can introduce complexity and performance overheads. Techniques such as homomorphic encryption are computationally intensive, which can affect the speed and efficiency of automated systems. Balancing privacy and performance is a critical consideration in the design of PET-enabled automation.

Integrating PETs into existing automation frameworks requires careful planning and expertise. Usability challenges may arise, particularly in ensuring that PETs are seamlessly incorporated into automated workflows without disrupting their functionality. Developing user-friendly tools and frameworks for PET integration is an ongoing focus for researchers and practitioners.

As privacy-enhancing technologies evolve, so do the methods employed by adversaries to compromise data. Continuous innovation and vigilance are required to stay ahead of emerging threats and ensure that PETs remain effective in protecting data privacy within automated systems.

Looking to the future

Efforts to standardise PETs are crucial for their widespread adoption and interoperability. Industry standards can facilitate the integration of PETs into diverse automated systems, ensuring consistency and reliability in data protection.

Ongoing research is driving advancements in PETs, making them more efficient and easier to implement. Innovations in areas such as lightweight homomorphic encryption and scalable differential privacy are expected to enhance the performance and applicability of PETs in automated environments.

The convergence of PETs with emerging technologies such as artificial intelligence and blockchain is poised to unlock new possibilities for privacy-enhanced automation. This integration could lead to more robust, secure and transparent automated systems that leverage the strengths of multiple technologies.

Privacy-enhancing technologies are reshaping the landscape of automation by enabling secure, data-driven decision-making while protecting sensitive information. As organisations continue to embrace automation, the integration of PETs will be essential in addressing privacy concerns and building trust among users. By balancing the need for data utilisation with the imperative of privacy protection, PETs pave the way for a future where automation can flourish without compromising individual rights or data security.

By Jeremy Bradley

Jeremy Bradley is the chief operating officer at Zama. He is a cross-functional and highly tactical leader who has worked with many organisations to shape strategy, drive communications and partnerships, and lead policy and process.

Find out how emerging tech trends are transforming tomorrow with our new podcast, Future Human: The Series. Listen now on Spotify, on Apple or wherever you get your podcasts.