Amazon OpenSearch Service now offers agentic AI capabilities that enable engineering and support teams to analyze log data through an agentic conversational interface. These agentic AI features help simplify log querying and accelerate incident investigations by allowing teams to interact with data using natural language, plan and initiate autonomous root cause analysis, and persist conversation as they navigate through their Observability workspace in OpenSearch UI.

This launch introduces three key capabilities available at no additional cost (token-based usage limits apply). Agentic chat enables you to ask questions in natural language to analyze data, generate and iterate Piped Processing Language (PPL) queries in Discover, and analyze visualizations for insights. When deeper root cause analysis is needed, you can trigger the investigation agent to autonomously and iteratively plan for the investigation, execute queries, reflect on results, and then deliver structured root cause hypotheses ranked by likelihood with full transparency into its reasoning. With agent memory, you can seamlessly continue your conversation across different feature pages or in a new web session.

You can use the agentic AI features in the following AWS Regions: Asia Pacific (Tokyo), Asia Pacific (Sydney), Europe (Frankfurt), Europe (Stockholm), Europe (Spain), Europe (Ireland), US East (N. Virginia), US East (Ohio), and US West (Oregon).

To learn more, see Agentic AI in Amazon OpenSearch Service. For more information about Amazon OpenSearch Service, see the Amazon OpenSearch Service product page.