Confluent Targets Real-Time AI Data Gaps

Confluent has added new capabilities to Confluent Intelligence and Confluent Cloud to help organizations build and secure real-time AI applications. The updates focus on the data layer behind AI, including streaming operations, privacy controls, private cloud connectivity, and developer workflows. This is to make it easier for teams to move AI applications into production without losing control of sensitive data or real-time context.

AI projects are running into the same problem: the models may be ready, but the data is fragmented, exposed, or difficult to use safely. Confluent is positioning its streaming platform as the connective layer between historic data, real-time data, and AI applications. Agentic AI systems need a current, governed context to make useful decisions, especially in industries where security, privacy, and compliance shape what can actually be deployed.

The new capabilities include a managed Model Context Protocol server and Agent Skills, which let developers use natural language to build, manage, and debug streaming operations. Confluent is also adding automated PII detection and redaction in Flink SQL, so sensitive data can be protected directly in streaming pipelines without moving it elsewhere first. Azure Private Link support gives teams private connectivity to Azure-hosted services, including Azure OpenAI, Azure SQL, and Cosmos DB.

Confluent is trying to reduce the handoffs between data engineering, AI development, and security teams by bringing governance and workflow automation into the streaming layer. That could help organizations build AI applications that are faster to update, easier to govern, and less dependent on manual data movement before they reach production.