Dynatrace Acquires Arize AI to Bridge Application and AI Observability

A fundamental shift is underway in how enterprises monitor their software. Artificial intelligence is reshaping both what needs to be observed and how operations teams respond when issues arise. Legacy observability systems were designed for deterministic applications, relying on traditional telemetry like logs, metrics and traces. AI-powered systems behave unpredictably, generating variable outputs from similar inputs, while organizations increasingly demand that observability tools do more than flag problems—they must supply the context needed for diagnosis, remediation and automated action.

Dynatrace's acquisition of Arize AI exemplifies this convergence. The deal brings AI observability, evaluation and agent monitoring into Dynatrace's application observability platform. During an appearance on theCUBE Research's AppDevANGLE podcast, Steve Tack, chief product officer of Dynatrace, and Aparna Dhinakaran, co-founder and chief product officer of Arize AI, discussed why application and AI observability are merging and what this means for enterprise operations. Tack remarked, "The world has shifted so much. AI brings new problems, new domains to the space."

From deterministic software to nondeterministic systems

Applications themselves are changing in fundamental ways. Conventional software delivers predictable results, but AI-driven systems—especially those using large language models and autonomous agents—introduce nondeterministic behavior that complicates troubleshooting. Observability can no longer stop at checking whether an application is running or infrastructure meets performance targets. Teams must verify that AI systems produce intended responses and that response quality meets standards.

Dhinakaran explained the shift: "Evaluating no longer just becomes about is it right or wrong. It becomes about actually measuring the quality of the responses, which is just a very fundamentally different problem." Arize built its platform to address this exact challenge, offering tools for tracing, evaluating and improving AI applications and agents. The company's open-source Phoenix platform serves more than 4,000 enterprises, while Arize AX provides a managed service for teams running AI systems in production. For Dynatrace, these capabilities extend observability into an increasingly critical application layer as enterprises move AI from pilots to production.

Observability brings shared context across AI and application telemetry

AI applications seldom operate in isolation. Agents invoke application programming interfaces, query databases, depend on cloud infrastructure and integrate with enterprise systems. When investigating AI behavior, teams face a challenge: that behavior is just one piece of a much larger software ecosystem. Arize customers increasingly sought stronger integration between AI telemetry and traditional application and production telemetry. Dynatrace customers, conversely, requested deeper AI observability and evaluation. Merging these two domains could provide developers, site reliability engineers, platform teams, AI engineers and data scientists with a unified perspective across the entire application stack.

Dhinakaran stated, "The agent systems and the software systems are joined at the hip. Having this ability to not only debug agents with AI observability, but also have all the context of the software that they use to call tools or the underlying infra behind the agents … just makes us build better products." This unified view also tackles another persistent observability challenge: tool proliferation. Research cited during the conversation showed that 75% of organizations deploy between six and 15 observability tools. As enterprises add AI monitoring, evaluation and governance systems, there is a risk that AI becomes another isolated operational silo rather than simplifying complexity.

Tack contended that merging application and AI observability delivers a more comprehensive system-level perspective rather than forcing teams to stitch together data from separate platforms. He said, "The real loss often happens [when] they lose the ability to have a system mindset. How can we bring a broader view together? How can we have shared context? How can we take action?"

From dashboards to operational action

The transformation may be less about what observability monitors and more about who—or what—uses that information. Historically, observability was built for engineers reviewing dashboards, responding to alerts and manually resolving incidents. AI agents introduce a different operational model where telemetry becomes input that software agents themselves can process.

Dhinakaran observed, "Observability is no longer about humans looking at dashboards and metrics and logs. It's about action." This reframes the purpose of observability data. Rather than merely documenting what occurred, telemetry can become part of the reasoning framework agents use to spot problems, suggest changes or trigger fixes. This shift also raises the bar for accuracy and context. Autonomous operations depend on organizations trusting the information driving those decisions.

Tack emphasized that delivering precise analytics and dependable answers will prove critical as enterprises grant agents greater authority. He stated, "How can we help them act, helping them move faster, creates so much opportunity." Dynatrace has been advancing this direction through its AI and automation initiatives, including Dynatrace Intelligence and BlueBox AI for agentic development and SRE workflows. Arize contributes deeper evaluation and observability for the AI systems participating in those workflows.

AI changes how software teams operate

The acquisition also reflects broader transformation in software development itself. AI agents are increasingly deployed not only within applications but to build, test, troubleshoot and operate them. Tack envisioned a future where architects spend less time directly in development environments and more time directing groups of specialized agents. Observability becomes part of the feedback mechanism connecting autonomous development to production operations.

Tack remarked, "The market's not just layering another technology on top. They are changing the way they want humans to work. Where does the agent step in?" For enterprises, this could eventually position observability as an operational intelligence layer spanning application development, AI evaluation, infrastructure and automated remediation. The challenge lies in advancing automation in tandem with the governance, reliability and confidence enterprises need before delegating significant operational decisions to agents.

The bottom line

Dynatrace's acquisition of Arize AI reflects two concurrent transformations: enterprise applications are becoming less deterministic, and observability is becoming more action-oriented. AI applications demand new approaches to evaluating behavior and quality, while AI agents increasingly require application and infrastructure context to make sound operational decisions. Integrating these telemetry environments positions Dynatrace to move beyond conventional application monitoring toward a model centered on shared context, evaluation and automation.

For developers and platform teams, the implication is significant: observability may increasingly serve as machine-consumable infrastructure. Dashboards will persist, but next-generation observability platforms must support both engineers diagnosing systems and agents helping operate them. As Dhinakaran concluded, "Every business is going to become an AI company," and tools for understanding and enhancing those agents will become "a core part of every stack."

https://www.youtube.com/embed/KccdVd-rJ88?feature=oembed