Analyst(s): Mitch Ashley
Publication Date: January 22, 2026
Harness introduced the Human-Aware Change Agent, a new AI system that integrates human conversation with automated change investigation during incidents. The agent listens to team dialogue, correlates observations with deployments, feature flags, and infrastructure changes, and surfaces evidence-backed hypotheses for faster resolution. This positions Harness to compete across incident management, extending beyond its core delivery automation platform.
What is Covered in this Article:
- Harness launched the Human-Aware Change Agent as part of its AI SRE suite, designed to capture human insight during incidents and connect it to change data across deployments, feature flags, configuration, and infrastructure.
- The agent uses AI Scribe to listen to team conversations in Slack, Teams, and Zoom, filtering operational signals and converting them into investigation actions that correlate human observations with system changes.
- Harness positions this as a new AI system designed to treat human insight as operational data during incident investigation, combining conversational context with what the company calls change intelligence across its Software Delivery Knowledge Graph.
- The announcement extends Harness’s platform from change automation (deployments, feature flags, pipelines) into agent-driven incident response, overlapping with observability and incident management vendors.
The News: Harness announced the Human-Aware Change Agent as part of its AI SRE incident response system. The agent uses AI Scribe to capture operational signals from team conversations in Slack, Teams, and Zoom, then correlates those observations with what Harness describes as change intelligence across its Software Delivery Knowledge Graph—deployments, feature flags, configuration, and infrastructure updates. The system surfaces evidence-backed hypotheses connecting what teams observe to what changed in production. For example, the agent might identify that a deployment introduced a new retry configuration 12 minutes before latency spiked, with supporting data showing downstream timeouts increased immediately afterward.
Harness positions this as a new AI system designed to treat human insight as operational data for incident investigation, creating, the company says, a unified approach in which conversational context and change intelligence inform each other rather than operate separately.
Harness Incident Agent: Is DevOps Now The AI Engineers of Software Delivery?
Analyst Take: Agents Across the Software Delivery Lifecycle — Through 2026, AI agents will emerge across operations, software delivery, CI/CD, software security, and QA. These agents will perform multi-step execution with guardrails, identity, and oversight structures embedded in delivery workflows, not function as assistants providing suggestions.
Just as developers are becoming engineers of agent-driven development, DevOps and platform teams are becoming engineers of software delivery using AI agents. The work shifts from manual pipeline orchestration and incident correlation to designing how agents investigate, remediate, and enforce policy across the lifecycle. This creates execution obligations for vendors across incident management, observability, and delivery platforms. Vendors must demonstrate equivalent integration depth or risk fragmentation where teams manage agents across disconnected tools.
Harness is moving beyond change automation into agent-driven incident investigation, a strategic shift that redefines where Harness competes. The core innovation is treating human conversation as operational data.
Harness Advances End-to-End AI Software Delivery Platform Positioning
The Human-Aware Change Agent connects to broader platform consolidation, executed through 2025. In Q3 2025, Harness simplified deployment strategies to checkbox-level configuration, strengthened GitOps integration with drift detection and unified rollback flows, and launched IDP 2.0 with bi-directional Git sync and Catalog Auto-Discovery.
These moves create a unified control plane where developers provision resources through IDP, platform teams orchestrate deployments through CD and GitOps, and operations teams investigate incidents through AI SRE.
The strategic signal is platform vendors competing to own workflow authority across build, deploy, and operate phases, reducing reliance on separate point tools. This creates execution pressure on delivery platforms like GitLab and incident response vendors like PagerDuty to demonstrate equivalent integration depth or risk fragmentation where teams manage disconnected agents across the lifecycle.
Competitive Structure Implications
The Human-Aware Change Agent creates competitive tension by extending beyond alerting and dashboards into agent-driven investigation. Harness is betting that change data, combined with conversational context, provides faster resolution than observability-first approaches that rely solely on logs and metrics.
The power dynamic shifts if Harness proves faster incident resolution through change-centric investigation. Observability vendors like Datadog and New Relic lose positioning if teams resolve incidents without needing to correlate logs and metrics manually. Incident management platforms like PagerDuty risk displacement if alerting and escalation become commoditized features within delivery platforms that already own change data.
The risk for Harness is execution depth. Incident response requires deep integration with observability, ticketing, and communication systems. The validation need is whether those integrations provide sufficient context for the agent to perform change investigation reliably, or whether teams still need to correlate data manually across systems. Customer proof at scale will determine whether this is a unified intelligence engine or a well-integrated assistant.
What Enterprises Demand
Futurum Intelligence data shows 78% of CIOs cite security and compliance as barriers to AI adoption, while only 8.8% of enterprises prioritize agentic features as top selection criteria when evaluating tools.
Harness must demonstrate that the Human-Aware Change Agent delivers measurable ROI (Futurum Intelligence shows 63% of organizations deploying AI report improved operational ROI) while addressing governance concerns that have stalled agent adoption elsewhere.
The agent’s ability to generate audit trails and evidence logs becomes critical not just for operational trust but for clearing organizational compliance hurdles. The test is whether those insights remain accurate under operational load, complex multi-service failures, and ambiguous conversational input. Agents that perform well in controlled demos often degrade when faced with real-world incident variability.
What to Watch:
- Customer validation of whether Harness’s change intelligence across deployments, flags, configuration, and infrastructure provides sufficient investigation scope, or whether teams still need observability data to correlate symptoms with root cause.
- Emerging evidence of measurable MTTR reduction and faster time to production recovery from enterprise deployments using Harness AI SRE.
- Observability vendors like Datadog and New Relic respond with agent-driven investigation capabilities that integrate logs, metrics, and traces with conversational context, creating competitive pressure on Harness’s change-centric approach.
- Incident management platforms like PagerDuty expand beyond alerting and escalation into AI-driven root cause analysis, overlapping directly with Harness AI SRE’s positioning.
- Guardrail and policy enforcement evidence showing how Harness prevents agent-driven investigation from surfacing incorrect hypotheses or recommending unsafe remediation actions.
See the blog post announcement on the Harness website for more information.
Disclosure: Futurum is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of Futurum as a whole.
Other insights from Futurum:
GitLab’s Salvo in the Agent Control Plane Race
Dynatrace Brings Feature Management Into the Observability Control Plane
Can Red Hat and NVIDIA Remove the Friction Slowing AI Deployments?
Author Information
Mitch Ashley is VP and Practice Lead of Software Lifecycle Engineering for The Futurum Group. Mitch has over 30+ years of experience as an entrepreneur, industry analyst, product development, and IT leader, with expertise in software engineering, cybersecurity, DevOps, DevSecOps, cloud, and AI. As an entrepreneur, CTO, CIO, and head of engineering, Mitch led the creation of award-winning cybersecurity products utilized in the private and public sectors, including the U.S. Department of Defense and all military branches. Mitch also led managed PKI services for broadband, Wi-Fi, IoT, energy management and 5G industries, product certification test labs, an online SaaS (93m transactions annually), and the development of video-on-demand and Internet cable services, and a national broadband network.
Mitch shares his experiences as an analyst, keynote and conference speaker, panelist, host, moderator, and expert interviewer discussing CIO/CTO leadership, product and software development, DevOps, DevSecOps, containerization, container orchestration, AI/ML/GenAI, platform engineering, SRE, and cybersecurity. He publishes his research on futurumgroup.com and TechstrongResearch.com/resources. He hosts multiple award-winning video and podcast series, including DevOps Unbound, CISO Talk, and Techstrong Gang.
