SAS Data Maker

SAS Data Maker is a synthetic data platform from SAS, offered on Microsoft Azure and sold through Microsoft Marketplace. It generates data that mirrors real-world data sets, either by augmenting existing data or by creating new data from existing sources or user-defined parameters. SAS says it is designed so teams can fuel AI models without exposing personally identifiable information.

What It Is

SAS Data Maker is a low-code/no-code synthetic data generation tool. According to SAS, it uses SAS algorithms to create synthetic data sets that replicate the patterns and relationships of the source data without using actual personal or sensitive details. It is positioned for AI training within the SAS Viya ecosystem and aims to reduce the cost of acquiring or collecting real-world data.

How It Works and What It Produces

Users generate synthetic data from existing data or defined parameters, then review results through automated visual statistical evaluation metrics that audit the quality, realism and privacy of the generated data. The output is privacy-preserving synthetic data that SAS says supports compliance with regulations such as GDPR and HIPAA. It can also generate additional samples for underrepresented classes to balance data distributions.

Documented Use Cases

SAS lists the following scenarios:

• Working around privacy restrictions on sensitive data
• Balancing imbalanced data sets to reduce model bias
• Creating data for rare events such as equipment failures
• Filling gaps where real data is sparse
• Fraud detection model training
• Drug discovery and clinical trial research
• Policy development
• Predictive maintenance
• Sales and pricing forecasting

SAS's World Wildlife Fund case describes using generated data to simulate donor behaviors and segments, which are then brought into SAS Customer Intelligence 360.

Availability

SAS states that SAS Data Maker is available globally through Microsoft Marketplace.

Purchase and deployment costs

The official SAS listing on Microsoft Marketplace specifies a $0 monthly control-plane base fee plus $1 per 1,000 synthetic rows generated. After purchase, a linking code connects a publisher-managed application deployed into the customer’s Azure tenant. Azure infrastructure charges are additional. The listing currently advertises a limited-time first-100,000-rows-free offer; eligibility and duration should be checked before purchase.