AWS Announces Knowledge Graph Service for AI Agents

Amazon's cloud service AWS announced three new services to construct a 'context layer' that enables AI agents to leverage enterprise data. The announcement was made at AWS Summit NYC in New York, where the flagship service 'AWS Context' was introduced alongside the general availability of 'Amazon S3 Annotations' and a preview of the skill asset feature for 'AWS Glue Data Catalog'.

The context layer is a mechanism that sits between large volumes of data held by an enterprise and an AI agent, enabling the agent to understand 'what it should reference.' Until now, this domain lacked standardized services, requiring each company to build its own implementations. AWS adopted an approach that automatically learns from agent usage patterns to address this challenge.

AWS Context is a service that automatically constructs knowledge graphs from an enterprise's existing data. A knowledge graph organizes the relationships between data like a map, automatically inferring information about which tables contain what information and what relationships exist between different data sources. Swami Sivasubramanian, Vice President of Agentic AI at AWS, explained: 'As the agent continues to be used, it learns which data sources deliver accurate results, and the knowledge graph itself improves automatically.'

Data administrators can review the inferred relationships in the AWS management console, apply business definitions and usage rules, and then deploy them to production. Access control leverages AWS's existing IAM and Lake Formation mechanisms, allowing tracking of who accessed which data. Metadata is stored in Amazon S3 in Apache Iceberg format, making it referenceable from standard engines such as Athena and Redshift, with a design that avoids dependency on AWS-proprietary APIs.

Amazon S3 Annotations, announced alongside these services, enables users to directly attach business meaning and supplementary information to individual files in storage. This allows files to carry context information at the point of storage, improving accuracy when AI agents reference them. The skill asset feature for AWS Glue Data Catalog is currently in preview, with detailed information to be disclosed at a later date.

The establishment of context layers has become a competitive field with participation from multiple vendors. AWS entered the market with a proprietary approach: 'a graph that learns automatically from agent usage without requiring manual data reorganization.' As agentic AI adoption expands, the ability to automate data preparation costs and efforts for enterprises becomes a critical factor determining future market penetration.