AWS released a semantic layer for agentic AI on AWS, using Denodo and Amazon Bedrock AgentCore, saying it enables governed access to live enterprise data across multiple sources. It is the company's first major update to its agentic AI platform since the introduction of Amazon Bedrock AgentCore in 2023.
The new semantic layer allows federated access to data sources such as Amazon Aurora, Amazon Redshift, and on-premises PostgreSQL databases, measured by its ability to support complex queries that span multiple systems. That compares with the previous semantic layer, which used Stardog’s knowledge graph and was limited to AWS data sources.
The semantic layer is built on Denodo’s Virtual DataPort and targets agentic AI applications that require real-time data access. Availability begins with a public preview, initially for enterprise customers using AWS and Denodo.
"The semantic layer provides a unified view of enterprise data, making it easier for agents to access and query data from multiple sources," said Denodo’s Chief Technology Officer, Carlos Sánchez. The layer also supports dynamic security policies that enforce row and column filters at query time.
The announcement follows the release of Amazon Bedrock AgentCore, which AWS said enables developers to build, connect, and optimize agents at scale. AWS did not say when the semantic layer will be fully available, and noted that it will continue to expand support for additional data sources.
Source: awsml
