Cross border and enterprise data sharing with synthetic data

Data is getting increasingly difficult to acquire, even within the walls of the same organization. Enterprise data sharing has long been a difficult process. Endless bureaucracy and suboptimal data outcomes make the lives of engineers and data scientists difficult. Off-shore development teams rely heavily on data sharing for testing applications. In-house operations need data sharing for scale. For example, analytics projects need to span several countries or continents. With data privacy legislations and rulings, like Schrems II effectively prohibiting US-EU data sharing, such projects get the axe before they would even take off. Similar data privacy regulations are popping up all over the world. An increasingly hostile cybersecurity environment further inhibits free data flows, even within the walls of heavily protected organizations. Cross border data sharing is getting increasingly difficult all over the world and the tide is unlikely to turn. 

Organizations, especially those handling troves of sensitive data, like financial institutions, banks and insurance companies, rely heavily on legacy data anonymization tools that hinder both privacy and data utility. Less mature organizations take unacceptable levels of risk. Using production data in non-production environments, such as testing should be a thing of the past no matter the industry.

McKinsey estimates that privacy-safe data sharing will generate almost $3 trillion annual economic value. Personal data sharing is off-limits, but synthetic data generators are here to help. AI-generated synthetic data is modeled on original data. Synthetic datasets or databases function as anonymous, yet meaningful drop-in placements for production data. Synthetic data does not qualify as personal data. As a result, it is out of scope for privacy laws, like GDPR. What’s more, high quality synthetic data is statistically identical to the original dataset or database it was modeled on. As a result, synthetic data can be used for application testing, data intensive POCs, cross-border analytics and AI/ML projects or to share with researchers and regulators. Synthetic data sandboxes are great data sharing tools, tried and tested in highly regulated environments from banking to insurance and healthcare.