aws AWS What's New ·

AWS Context Ontology Accelerator introduced to aid AI decision-making

aiawspreviewengineeraws-bedrock
feature announcement

AWS has launched the open-source Context Ontology Accelerator to help organizations build machine-readable business ontologies, improving AI agent decision accuracy and explainability. This tool connects to data sources and uses AI to draft ontologies, which domain experts then review and approve, storing them in a knowledge graph. The accelerator aims to drastically reduce the time and effort required for ontology authoring, moving AI agents from proof-of-concept to production with greater trust and auditability.

  • Open-source Context Ontology Accelerator aids AI agent decision-making
  • Connects to diverse data sources for ontology drafting
  • Ontologies stored in knowledge graphs and accessible via standard protocols
  • Reduces ontology authoring time from months to days
  • Initial release integrates with AWS services and is available on GitHub
Features (1)
  • Open-source Context Ontology Accelerator aids AI agent decision-making

    AWS introduces the Context Ontology Accelerator, an open-source tool that helps organizations create machine-readable ontologies of their business. This enables AI agents to make more accurate, consistent, explainable, and auditable decisions by providing them with structured business context.

Enhancements (3)
  • Connects to diverse data sources for ontology drafting

    The accelerator connects to both structured and unstructured data sources, leveraging AI to draft an initial ontology. Domain experts then review, edit, and approve each element, ensuring accuracy and control over the resulting knowledge graph.

  • Ontologies stored in knowledge graphs and accessible via standard protocols

    Approved ontologies are stored in a user-owned knowledge graph using open W3C standards, ensuring compatibility with various tools. Any agent can consume this information through the provided Model Context Protocol (MCP) server.

  • Future integration with AWS Context planned

    The managed, user-defined ontology capability of the Context Ontology Accelerator is slated to become a fully managed feature within AWS Context, allowing customers to use and manage their created ontologies directly with the service.

Notes (2)
  • Reduces ontology authoring time from months to days

    Context Ontology Accelerator significantly decreases the manual effort and time required for ontology authoring, aiming to streamline the process from data connection to serving governed answers to AI agents.

  • Initial release integrates with AWS services and is available on GitHub

    The accelerator is available under Apache 2.0 on GitHub, initially using Amazon Neptune for graph storage, Amazon OpenSearch Serverless for vector storage, and foundation models from Amazon Bedrock. Deployment can be done independently or with AWS Professional Services.

Read the original announcement →

https://aws.amazon.com/about-aws/whats-new/2026/07/aws-context--ontology-accelarator-generally-available

Related releases