New Relic launches AI knowledge layer for IT incidents
Wed, 6th May 2026 (Today)
New Relic has launched New Relic Knowledge, a feature aimed at customers using AI in IT operations.
The feature combines real-time telemetry with historical incident records, system changes and operational data to help engineers and AI agents identify and resolve technical issues. It sits across the broader New Relic observability platform as a layer intended to give systems more context during incident detection and response.
Businesses are pushing more operational tasks towards automation, but autonomous systems still depend on accurate, current data on how software and infrastructure are behaving. New Relic is positioning the product around that challenge, arguing that AI tools need access to institutional knowledge as well as live system signals to make reliable decisions.
The product is intended for Site Reliability Engineering teams, DevOps staff and platform engineers, who often move between dashboards, logs, documents and ticketing systems to understand what has gone wrong. New Relic says engineers can spend up to 40% of their time searching for context across those sources.
How it works
New Relic Knowledge analyses telemetry data across metrics, logs, traces and events, then correlates it with prior incidents, service relationships and recent changes such as deployments or configuration updates. The goal is to surface explanations more quickly after an alert appears, reducing the manual investigation needed before teams can begin remediation.
The feature can also be used by AI agents to diagnose incidents and suggest next steps by referencing similar past problems and behavioural patterns across systems. That reflects a broader shift in the observability market, where vendors are trying to move beyond monitoring dashboards into tools that support automated triage and response.
New Relic linked the launch to the financial cost of outages, citing a median annual downtime risk of USD $76 million for organisations. Its argument is that faster resolution can reduce that exposure by cutting mean time to resolution, the metric IT teams use to measure how long it takes to restore service after an incident.
Brian Emerson, chief product officer at New Relic, outlined the company's rationale for the product.
"Organisations today must solve technology problems at a pace that far exceeds human scale. While AI agents are addressing this challenge, they are only as effective as the data they can access," said Emerson. "New Relic Knowledge provides the connective tissue between telemetry and action, ensuring that every technical decision-whether made by a human or an agent-is grounded in real-world context to drive true business impact."
Market shift
The launch comes as software suppliers across infrastructure management, application monitoring and security operations seek to package internal data and historical records into forms that generative AI systems can query. For many of those efforts, the central challenge is not simply access to live data, but combining it with enough organisational and technical context to avoid false conclusions.
In practice, that means observability providers are trying to connect logs and metrics with change records, service maps, deployment history and prior incident reports. New Relic's latest addition follows that pattern by attempting to turn fragmented operational data into a more unified knowledge layer for both people and software agents.
New Relic describes the product as a continuous intelligence layer rather than a static knowledge base. In operational terms, that means it is intended to respond to user queries and incident conditions using current telemetry and stored business information, rather than relying only on manually maintained documentation.
For customers, the value will depend on how accurately the system can connect anomalies with relevant historical events and whether that reduces the workload on engineering teams during outages. Vendors in this segment have increasingly focused on lowering operational toil, especially in environments where applications, cloud services and infrastructure are spread across multiple tools and teams.
New Relic Knowledge will be generally available to New Relic AI customers on May 25, 2026.