Feature overview
AI Assistant in Observability Cloud is a GenAI-powered experience that empowers engineers to find and fix issues faster using natural language. Using an agentic framework powered by an LLM (Large Language Model) and knowledge bases, the AI Assistant analyzes your environment, unearths new insights, and provides workflow guidance to accelerate your investigations, regardless of your experience in Splunk or observability.
AI Assistant in Observability Cloud leverages an agentic framework, where a versatile LLM is enhanced with tool integrations and knowledge bases to address user queries more effectively. The agentic framework refers to an intelligent agent that can orchestrate domain-specific behavior based on user queries. At its core, an orchestration agent interprets user intent, plans workflows, invokes the right tools, and synthesizes troubleshooting insights. Within Splunk Observability Cloud, this orchestrator interacts with services like Splunk Application Performance Monitoring (APM), Infrastructure Monitoring, and Log Observer Connect to enable more efficient troubleshooting and insight generation tailored to your environment.
Model overview
The AI Assistant in Observability Cloud is powered by a suite of cutting-edge LLMs provided by a trusted cloud vendor, helping to drive more seamless integration and better performance. This ensemble of LLMs is categorized into three broad categories: an orchestration model that expertly coordinates all tasks; domain-specific models equipped with knowledge bases to tackle complex tasks; and embedding models that facilitate more efficient knowledge indexing and retrieval.
Model evaluation and performance
AI Assistant in Observability Cloud is evaluated with an end-to-end protocol: the assistant is given a diverse test set of questions that are representative of the questions that our users are likely to ask, and its answers are compared with the correct answers in the Splunk Observability Cloud product. Through iterative refinements to prompt engineering, tool description, and workflow design, the AI Assistant demonstrated its potential for driving value in real-world scenarios.
Data sources for model training
Domain-specific internal data sources are used to build the knowledge bases. Splunk uses information from customers’ interaction with AI Assistant in Observability Cloud in accordance with Specific Terms for Splunk Offerings and Documentation.
Data privacy and security
Data that is directly relevant to the generation of a meaningful response to a user prompt (such as the user prompt itself, tool description, and the data analytic results from the tools) are sent to the third-party LLM services that, in turn, generate responses. Data that is not necessary to generate a relevant response stays within the Splunk compliance boundary for the customer’s underlying Splunk Cloud Platform offering. Data categories relevant to Splunk’s AI Assistant are described in Data sharing and use. Additionally, the Specific Terms for Splunk Offerings set forth Splunk’s data practices relating to all data collected, generated, or used by AI Assistant, including our practices relating to your opt-out preferences.
