The AI incident assistant is a chat feature embedded in the incident page. It streamlines all incident context—including
alerts, descriptions, and impacted topology—to the LLM, helping on-call engineers gather information faster and
resolve incidents more efficiently. Users can ask for root cause analysis and even execute commands on third-party
services (read more about provider methods).
Model used: OpenAI, a model hosted by Keep, or other. Data flow: Data is shared between LLM provider and Keep whether the LLM provider may vary depending on the contract.