Facing an unprecedented wave of software failures driven by artificial intelligence (AI), cloud computing provider PagerDuty Inc. on Tuesday unveiled Paige, an autonomous Site Reliability Engineering (SRE) agent designed to diagnose and repair technical disruptions alongside human engineers.
The announcement comes as enterprises confront a structural shift in enterprise IT. Software is increasingly written, tested, and deployed by AI with minimal human oversight. While automated pipelines accelerate delivery cycles, they also introduce complex, hard-to-diagnose failure points into production environments, ranging from model drift and token runaway to data leakage and hallucinations.
According to industry data cited by the company, 95% of corporate executives view incident reduction as a primary competitive advantage, with 77% intending to ramp up resilience spending over the coming year.
“A tsunami of fragile AI-generated code is now entering production, and SRE teams are stretched to the max,” PagerDuty CEO John DiLullo said in a statement. “This new code breaks differently and more often, and practitioners find it increasingly difficult to determine the root cause when a problem does arise. The only way to maintain resilience in the face of this complexity is to fight fire with fire.”
Paige acts as a virtual teammate embedded directly within operations and DevOps organizations. The agent offers flexible operational modes. It can execute automated remediation for standard, well-documented disruptions, or function in a human-in-the-loop posture where it conducts initial telemetry triage while leaving final approval to engineers. Operating on an adaptive learning architecture, the system archives context from every resolved event into a shared operational memory to sharpen future troubleshooting.
The agent operates atop the newly rebranded PD Reliability Platform — formerly known as the PagerDuty Operations Cloud — which consolidates disparate telemetry feeds, incident workflows, and automation scripts into a centralized command center.
Integrated natively into enterprise chat software such as Slack and Microsoft Teams, the platform automatically provisions conference bridges, drafts post-incident reviews, and populates follow-up action items with assigned owners upon resolution.
To keep pace with the swift evolution of generative workloads, the company also launched PD Labs, an external collaborative incubator where enterprise design partners can test experimental resilience prototypes against live production challenges.
Industry analysts observe that automated interventions are shifting from convenience to structural necessity.
“IT teams are looking for value from their AI agent investments, including measurable outcomes such as improved uptime, lower cost of operations, and time returned to responders,” noted Snow Tempest, research manager for IT service management at IDC.
PagerDuty, which manages nearly one billion technical disruptions annually and serves two-thirds of the Fortune 100, is making the platform available via a single scale-tiered fee that bundles telemetry event ingestion, automation actions, and user seats into a unified deployment model.

