AWS Context boosts AI reasoning with governed knowledge graphs

AWS has introduced a new service called AWS Context. It maps a company’s data relationships into a knowledge graph for AI agents. The tool aims to improve AI reasoning by providing governed, structured context at runtime. AWS Context allows AI agents to access business rules, domain knowledge, and data relationships securely. Each query is identity-aware, ensuring agents only retrieve data their identity permits. All interactions remain auditable for compliance and oversight.

The service aggregates context across different information silos using structural and semantic traversal. Developers can control which datasets are included in the knowledge graph, excluding irrelevant or sensitive information. Over time, AWS Context learns from usage patterns to refine its responses.

Mai-Lan Tomsen Bukovec, AWS vice president of technology for data and analytics, described the service as a 'data lake of nuance and information'. It provides AI agents with a shared, governed context layer to enhance decision-making. AWS also announced a preview of new business context and semantic search functions for AWS Glue Data Catalog, simplifying data discovery for both humans and AI agents. AWS Context strengthens AI reasoning by delivering curated, identity-aware insights. The service integrates with existing data repositories and improves with use. It offers a scalable way for organisations to build smarter, more transparent AI systems.