Amazon Web Services (AWS) recently unveiled AWS Context, a new service aimed at significantly enhancing the intelligence and reliability of AI agents within organizations. Announced at the AWS Summit New York City, AWS Context is designed to address the pervasive challenge of fragmented enterprise data by automatically mapping relationships across various disparate data sources—including data lakes, data warehouses, databases, and real-time streams—into a cohesive and actionable knowledge graph. This forthcoming service will provide AI agents with advanced "agentic search" capabilities, allowing them to access governed data relationships, critical business rules, and specialized domain-specific knowledge in real-time, thereby enabling more informed, accurate, and trustworthy decision-making across a wide range of applications.
The effectiveness and trustworthiness of AI agents are directly tied to their ability to access and reason over relevant, high-quality context. Currently, critical organizational information is often dispersed across numerous siloed systems and even exists as unwritten institutional knowledge, making it inherently difficult for AI agents to achieve optimal performance and reliability. AWS Context builds upon the same robust knowledge graph technology that already powers Amazon Quick, a service where hundreds of thousands of users interact daily with a production knowledge graph. By extending this proven personal knowledge graph concept into an organizational, shared, and governed context layer, AWS aims to provide a foundational solution for enterprises struggling with data silos and the complexities of integrating diverse information for sophisticated AI applications. This strategic move positions AWS to further solidify its role in providing essential infrastructure and tools for widespread enterprise AI adoption.
The introduction of AWS Context is expected to have substantial implications for enterprises deploying and managing AI agents, promising to significantly streamline the process of providing these agents with the necessary, accurate context. This will reduce the considerable manual effort often involved in data preparation, integration, and governance for AI initiatives. Data stewards and curators will gain an intuitive console experience to manage the knowledge graph, allowing them to review inferred relationships, promote them to production, and attach crucial domain-specific information like business definitions and usage rules. Existing Amazon Quick users will immediately benefit from access to this broader enterprise knowledge graph, enhancing their agents' capabilities with cross-system relationships and curated context. Furthermore, the service integrates seamlessly with key AWS offerings such as AWS Glue Data Catalog, Amazon SageMaker Unified Studio, and AWS Lake Formation, ensuring robust governance, permissions management, and automated context addition with AI assistance. By publishing key elements to Amazon S3 in Apache Iceberg format, AWS also offers flexibility for customers to use their preferred Iceberg-compliant tools, fostering an open and adaptable ecosystem for enterprise AI development and innovation.