Elastic has expanded its collaboration with OpenAI to help organisations build production AI applications that draw on their own enterprise data. For organisations that deploy the two together, Elasticsearch acts as the retrieval and governance layer feeding OpenAI's reasoning models.
The result, it says, is more accurate, secure and reliable AI at scale.
The companies argue that AI agents are only as useful as the enterprise context they can reach. Most of that context, they say, stays out of reach. Unstructured data such as documents and multimedia makes up the great majority of enterprise data. It is spread across documents, tickets, logs, metrics, traces and security alerts, changes constantly, and is governed by different permissions.
Elasticsearch provides the retrieval layer, combining lexical and vector search, semantic reranking, filtering and access controls in one platform. Elastic says this helps agents find relevant, permission-aware context, reduces unnecessary data sent to the model, and lowers token costs.
"The success of enterprise AI depends on connecting powerful models with the knowledge businesses already possess," said Greg Tademoto, a global vice president at Elastic. "Much of that knowledge is buried in unstructured data."
Colleen Kapase is vice president of strategic global partnerships and ecosystems at OpenAI. She said the collaboration would bring OpenAI's models closer to the data businesses rely on.
Security and observability
The companies will focus on three outcomes. The first is context-aware agents that retrieve permission-aware knowledge. The second is agentic observability that correlates telemetry for root cause investigation. The third is agentic security operations that turn high volumes of alerts into evidence-backed investigations.
Through the OpenAI Daybreak Cyber Partner Program, Elastic plans to integrate OpenAI's GPT-5.5 cyber-specific models into Elastic Security workflows. It will also extend governance to the OpenAI platform. That enables detection of anomalous OpenAI activity alongside endpoint and network threats. For developers using OpenAI Codex, Elastic will build integration points that provide governed access to unstructured enterprise data.
Elastic has supported OpenAI models through AI Assistants and connectors since 2023. It says developers can begin connecting OpenAI models with Elasticsearch today using Elastic AI Assistants, Agent Builder and MCP integrations. Further details are set out in Elastic's blog. https://www.elastic.co