Ant International and Nanyang Technological University, Singapore (NTU Singapore) with an aim to strengthen the digital economic growth in the Asia Pacific region have announced collaboration over the next 5 years for conducting breakthrough research to advance digital trust by leveraging NTU’s Singapore research expertise on Privacy Enhancing Technologies (PETs) and Ant International’s industry experience.
Mr Jerry Yin, Chief Technology Officer of Ant International, said: “Ant International is pleased to expand our collaboration with NTU, through this longer-term partnership. As a global digital payment and financial technology provider, data privacy is a core part of our business as we aim to provide merchants with secure, reliable and seamless solutions. By leveraging NTU’s academic expertise and Ant International’s industry experience, we look forward to advancing the development of privacy-enhancing technologies with new innovations that address real business needs.”
Privacy Enhancing Computation is revolutionizing the data privacy segment by rendering transformative solutions which balance the requirements of data utility with the imperative of safeguarding the personal information. Privacy Enhancing Computation technologies are emerging as one of the vital tools to ensure that data can be leveraged and shared securely, in an age where data breaches as well as privacy concerns are prevalent. In this article let us look at the various aspects of PEC, the technologies, applications & the way it is reshaping privacy norms in various industries.
PEC comprises a suite of technologies that are designed for protecting data privacy while allowing useful computations on either encrypted or anonymized data. These technologies have an aim to mitigate privacy risks which are associated with data processing as well as sharing and this enables enterprises to harness the potential of data without compromising individual privacy. Some of the key PEC technologies include Secure Multi-Party Computation, Federated learning, differential privacy, homomorphic encryption & zero-knowledge proofs.
Healthcare
The potential for PEC is vast in the healthcare industry. Medical research most often needs access to vast amounts of patient data. This is sensitive as well as protected under regulations such as HIPAA. Privacy Enhancing Computation technologies help researchers in analyzing encrypted patient data, deriving insights as well as training ML models, without exposing the personal health data. And this helps in enhancing privacy & promoting collaboration across institutions.
For instance, homomorphic encryption helps in the development of predictive models for various types of disease outbreaks through the analysis of encrypted health records from numerous hospitals. Also, federated learning can be implemented for enhancing diagnostic algorithms by training on decentralized data from numerous clinics which helps to ensure that patient data stays on local servers.
“The increased use of medical devices, bioprinting, robotics, predictive data analysis, precision medicine, and AI in drug discovery accelerates innovation but also intensifies the need for robust data privacy practices,” says Ganesh Nathella, Senior Vice President & General Manager – Healthcare and Life Sciences Business at Persistent Systems.
“AI/ML can enhance disease diagnosis and treatment personalization, but they require access to extensive patient data. Ensuring this data is anonymized and securely stored is paramount. Federated learning offers a novel approach to training machine learning models on decentralized data. This technique keeps raw data on individual devices, significantly reducing privacy risks associated with sharing sensitive information,” he adds.
Finance
As the financial industry deals with highly sensitive data, it makes it a prime beneficiary of Privacy Enhancing Computation. SMPC (Secure Multi-Party Computation) helps in fraud detection by enabling financial institutions in collaborating & sharing insights without revealing the data of customers. Therefore, this collaborative approach improves the detection of fraudulent patterns across various enterprises while maintaining privacy of data.
“One of the key technologies to watch is Differential Privacy. This approach allows one to compute some aggregated statistics without learning anything about the individual data points. This is particularly useful for web and mobile analytics, as you would be able to get cohort patterns without any information about what individual users are doing,” says Deeptech investor and CEO at homomorphic encryption startup Zama, Dr Rand Hindi.
Financial firms use differential privacy techniques for analyzing transaction data as well as detecting anomalies without compromising the individual privacy. This helps in building trust with customers by ensuring compliance with data protection regulations.
“While society harnesses the benefits of emerging technologies in the financial sector, regulators should pay careful attention to the underlying risks,” said RBI Deputy Governor MD Patra at a recent SAARCFINANCE seminar on “Emerging Digital Technologies in Central Banking and Finance” in Goa
Patra underscored that with the increasing use of AI, concerns arise about transparency, data biases, governance, privacy and the robustness of algorithms.
Government & Public Sector
Governments hold a humongous amount of data on citizens’ right from tax to social security information. Privacy Enhancing Computation facilitates sharing & analyzing this data across departments as well as agencies. For example, differential privacy can be utilized in the census process for providing accurate demographic data while ensuring that individual responses stay confidential.
Zero-knowledge proofs can be employed for secure & private voting systems. This ensures the integrity of voting process without exposing the identities as well as choices of the voters. This technology comprises applications in identity verification and this allows citizens for proving their eligibility for services without revealing personal information.
Business & Marketing
Businesses can leverage Privacy Enhancing Computation for enhancing customer trust & complying with data protection laws like General Data Protection Regulation (GDPR). Differential privacy & homomorphic encryption facilitates organizations in performing data analytics as well as derive consumer insights without accessing raw data. And, this approach helps businesses in personalizing marketing strategies while respecting customer privacy.
For training recommendation systems across various organizations federated learning can be implemented. This helps in rendering better product recommendations without sharing customer data between businesses and this collaborative approach helps in enhancing the quality of services while maintaining data confidentiality.
PEC is fundamentally changing the way we handle data privacy across industries. Privacy Enhancing Computation technologies facilitate in bridging the gap between privacy protection as well as data utility and this can be achieved by enabling secure computations on the encrypted or anonymized data. Privacy Enhancing Computation is helping in sharing of secure data & collaboration from healthcare & finance to government & business, fostering innovation while at the same time safeguarding individual privacy.
The future of Privacy Enhancing Computation looks bright & promising with the ongoing advancements in technology and thus increasing the privacy issues awareness. Privacy Enhancing Computation is likely to become an integral part of data management strategies across various sectors. Also, embracing such technologies not only helps in enhancing privacy but also helps in building trust and making sure that data can be leveraged responsibly.
