March 2022 Summaries

Corporations in regulated industries often fail to utilize a significant portion of their data due to privacy concerns, as data privacy legislation varies across regions and even within countries. Privacy Enhancing Technologies (PETs) offer a solution by enabling organizations to use sensitive data while maintaining privacy and security through methods like fully homomorphic encryption, multiparty computation, and federated learning. These technologies allow for secure data analysis without exposing raw data, facilitating collaboration on encrypted data while complying with data localization laws. PETs can be combined to enhance data value and protect against data breaches, with applications in fighting fraud, cyber threat intelligence sharing, and genome-wide association studies. By integrating PETs, businesses can maximize data utility without compromising privacy, paving the way for more informed decision-making.

The text highlights the experiences and aspirations of women working at Duality, a company focused on leveraging data science and cryptography to solve significant global challenges. Rina Shainski, Co-founder and Chairwoman, emphasizes her passion for technology-driven innovation, particularly in addressing conflicts between public good and data privacy, such as those encountered during the COVID-19 pandemic. Marcella Arthur, Vice President of Global Marketing, values mentorship and the role of technology in addressing world issues, while Rina Serlin, Vice President of People, discusses the importance of a diverse, global team in fostering innovation. Ruthi Franco, Director of Product Management, revels in the collaborative process of creating customer-centric products, and Saraswathy RV, a research scientist, highlights working on homomorphic encryption and privacy computation. Saroja Erabelli, a software developer, shares her enthusiasm for cryptography and its real-world applications, underscoring the company's mission to solve pressing privacy and security challenges.

As the volume of data grows exponentially, the challenge of maintaining privacy while extracting value becomes increasingly significant, leading to the rise of Privacy Enhancing Technologies (PETs). These technologies enable data scientists to analyze sensitive data without exposing it, ensuring compliance and privacy protection. The primary PETs discussed include homomorphic encryption, multiparty computation, differential privacy, federated learning, secure enclaves, and zero-knowledge proofs, each offering unique methods to process data securely. Additionally, synthetic and anonymized data are sometimes classified under PETs, though they come with the drawback of potential data devaluation due to the alteration of original inputs. These technologies, while varying in application and efficiency, collectively highlight the potential to unlock significant economic value from underutilized data by preserving privacy. As companies navigate these options, combining PETs can provide enhanced benefits, making them a strategic choice for maximizing data value while safeguarding privacy.