Data quality is the foundation of trustworthy analytics and intelligent applications. Azure SQL Database now brings AI-powered capabilities that make measuring and improving data quality smarter and more scalable. Using native vector data support and semantic similarity functions, teams can identify duplicate or inconsistent records beyond simple string matching. For example, a retail company can detect customer entries like “Jon Smith” and “Jonathan Smyth” as potential duplicates by comparing embeddings stored in SQL tables.

Integration with Azure OpenAI enables Retrieval-Augmented Generation (RAG) for context-aware validation—such as generating natural language summaries of anomalies in product descriptions.

Copilot for Azure SQL assists DBAs and developers by creating cleansing queries or recommending normalization steps using plain English prompts. Combined with hybrid semantic search through Azure AI Search, organizations can assess completeness and accuracy across millions of records quickly.

Improve data quality with AI in Azure SQL. Detect anomalies using embeddings, enrich missing details with RAG, and validate consistency through semantic checks. Copilot drafts schema constraints and cleanup queries, accelerating delivery and ensuring trusted data.

Not yet available, but posting here for future events