DataMasque listed as a solution provider on PathFin.ai

DataMasque unlocks production-realistic data for AI development and testing while keeping sensitive customer information protected.

AUCKLAND, New Zealand – 16 October 2026 – DataMasque, the sensitive data discovery and masking platform, today announced that it is now listed on the Monetary Authority of Singapore (MAS) Pathfinder Programme – PathFin.ai.

PathFin.ai is a collaborative initiative between MAS and the financial industry that fosters knowledge exchange in Artificial Intelligence (AI) implementations. PathFin.ai curates a library of AI solutions and implementation best practices among financial institutions. By centralising these resources and insights, the programme seeks to reduce the time and effort financial institutions need to search, select and effectively implement AI solutions.

Preserving data integrity without compromising security

Financial institutions cannot effectively train or fine-tune AI models without data that reflects real customer information. Models need realistic data to learn from the patterns, relationships and edge cases that are found in real-world environments. However, using actual customer data can introduce substantial regulatory and privacy risks.

To manage these risks, institutions have traditionally relied on approaches such as redaction, scrambling or dynamic masking, which tend to either strip out the detail models need to learn from, or leave sensitive information exposed.

DataMasque overcomes this challenge – the trade-off between data privacy and data utility – by replacing sensitive information with ‘synthetically identical’ data that replicates the format and referential integrity of real data, providing AI models with data they can learn from while ensuring no personally identifiable information ever leaves a secure production zone. 

This also makes federated learning viable for institutions building shared models – like a sector-wide financial crime model. Keeping sensitive records in-house is not enough since the model updates participants share can still reveal whether a particular individual was in the training data. If each participant trains on 'synthetically identical' records, what can be inferred from the model relates to synthetic identities rather than real customers, while consistent masking rules across institutions preserve the cross-organisational patterns the model needs to learn.

Case study: Mitigating regulatory exposure for an APAC bank

A major Asia-Pacific bank needed to migrate its on-premise SAS credit risk system to the cloud without violating regulations tied to exposed PII and sensitive data in non-production environments.

The bank implemented DataMasque in an isolated, air-gapped environment, where the platform combined metadata analysis, in-data discovery and a locally deployed, fine-tuned small language model to locate sensitive data across the bank's Hadoop storage.

DataMasque then generated ‘synthetically identical’ masked values – preserving exact formats and referential integrity – using one-way, irreversible SHA-512 salted hashing to prevent reverse engineering. The entire process ran within the bank’s own environment, so PII never left the secure zone.

The results:

• The bank’s open regulatory risk issue was closed, directly reducing exposure on the CIO’s risk register. 
• A third-party vendor successfully tested all credit risk pipelines — including risk-weighted assets and expected credit loss computations — against production-realistic masked data with full referential integrity, eliminating prior pipeline failures and rework. 
• Data refresh cycles are moving from annual to daily, weekly or monthly, significantly improving data currency for development and testing teams. 
• End users require only two to three hours of training to use the platform. 

The proof of concept ran for approximately six weeks, with full deployment completed within two months of commercial and regulatory approval.

“Most AI projects in financial services stall on data access," said Grant de Leeuw, Co-Founder and CEO of DataMasque. "Using real customer data carries unacceptable regulatory and privacy risks, and the usual alternatives either strip out the detail a model needs to learn from or leave sensitive information exposed. DataMasque removes that trade-off: institutions get synthetically identical data inside their own environment, with the realism and referential integrity their models need. As a listed PathFin.ai solution provider, DataMasque looks forward to supporting financial institutions exploring privacy-preserving AI adoption.”

About DataMasque

DataMasque is a sensitive data discovery and masking platform that enables enterprises to generate synthetically identical customer data for AI training, testing and analytics. Founded in New Zealand and operating globally, DataMasque serves regulated enterprises across financial services, insurance, hospitality and the public sector. The company is backed by Wavemaker Partners, OIF Ventures and Icehouse Ventures. For more information, visit DataMasque.com.

‍