Data streaming company Confluent (now an IBM company) confluent intelligence and confluence cloud.
This update is designed to streamline the way you build and secure real-time AI apps.
This new feature is said to remove barriers to production-ready AI applications with agent-driven workflows, automated data protection, and private cloud connectivity.
Confluent says it integrates the AI lifecycle with the tools developers already use, integrating Apache Flink pipelines with dbt, and introducing a fully managed Model Context Protocol (MCP) server and agent skills that allow AI to manage streaming operations.
Confluent embeds enterprise-grade governance directly into your data streams with automated Personally Identifiable Information (PII) redactions and private connectivity to external models via Azure Private Link.
“Most AI projects fail before reaching a single customer because the data layer breaks,” said Sean Falconer, Head of AI at Confluent. “Teams have models and permissions, but security risks and fragmented data prevent them from being released. We solve this problem by making the streaming layer the foundation for secure, production-ready AI.”
According to the McKinsey report, the problem is widespread. “…8 out of 10 companies cite data limitations as a barrier to scaling agent-based AI.”
Roots of data limitations
The root cause often has to do with security teams blocking data from entering AI pipelines due to leakage risks, and developers spending time switching between tools to inspect and manage the data streams that AI relies on. As a result, manual processes become slow and otherwise fast iterative cycles become bottlenecks.
The company proposes that Confluent Cloud and Confluent Intelligence form a production-ready data streaming foundation for AI that continuously processes historical and real-time data and provides it as trusted context to AI applications.
Developers can use Confluent MCP as a control plane, allowing AI to use natural language to build, manage, and debug streaming operations. Agent skills add a second layer of encoding best practices and workflows to ensure these operations are performed consistently and in line with your organization’s standards.
Together, they enable developers to use AI-powered tools to create and continuously improve real-time applications, bringing streaming into modern agent-driven development workflows. Generally available on Confluent Cloud.
automated data privacy
New built-in ML functions for PII discovery and redaction protect sensitive information directly in Flink SQL without the need for custom code, external services, or first moving data to a warehouse.
This will enable more AI use cases across highly regulated industries such as financial services, healthcare, and insurance. Available in early access for Confluent Intelligence.
support for Azure private link Ensure your AI workloads stay off the public internet with secure private paths to external model calls and external table queries. Flink jobs can now securely connect to Azure-hosted services such as Azure OpenAI, Azure SQL, and Cosmos DB over Microsoft’s private backbone. Generally available on Confluent Cloud.
Integrated engineering workflow
of Free open source DBT adapter Incorporate Flink SQL on Confluent Cloud into dbt, the industry-standard framework data that engineers use to build and manage data pipelines.
Teams can quickly define, test, and deploy streaming pipelines using the same dbt commands and project structure they use today.
This lowers the barrier to Flink adoption and makes it easier to extend existing data workflows to real-time use cases. Generally available on Confluent Cloud.
Confluent supports TimesFM models for robust anomaly detection and Anthropic and Fireworks AI models that developers can use directly in Flink stream processing workflows to build advanced real-time AI applications.
