AI coding agents are supposed to make software cheaper to build. They may also create more software changes, production failures and machine-generated complexity to inspect. That second-order effect connects Datadog, Inc. (NASDAQ:DDOG) and Dynatrace, Inc. (NYSE:DT) to an observability market that AI could expand rather than erase.
Datadog told the Goldman Sachs Communacopia + Technology Conference on September 10 that monetization of AI-agent observability has begun, with thousands of customers using the capability. Management described an emerging “inference economy”: more AI applications and agents mean more activity that customers need to monitor. That shifts the opportunity from watching human-written releases toward tracing continuous decisions made by software that can change software itself.
One day later, Dynatrace launched a verified Cursor Marketplace plugin. A single installation gives coding agents access to live production context through MCP and 30 Dynatrace skills. The timing makes the competition tangible: Datadog is framing the category, while Dynatrace is putting production telemetry directly inside an agent’s workflow.
Usage can grow faster than seats
Datadog has the stronger recent growth profile. Second-quarter revenue rose 36% to $1.12 billion, and customers generating at least $100,000 of annual recurring revenue increased 23% to roughly 4,720. Its usage-based model can capture expanding telemetry and inference volume. The risk is the same sensitivity to optimization that has periodically hurt cloud vendors; more capable agents could also resolve some incidents.
Dynatrace offers a different defense. It prices the Cursor integration on data consumption rather than per user, reducing direct exposure to shrinking developer-seat counts. Its latest quarterly ARR grew 17% to $2.136 billion, while annualized log-consumption reached $200 million after almost doubling over the previous two quarters. Slower companywide growth, plus the planned $915 million Arize acquisition, raises execution and integration risk.
Insider Monkey’s hedge fund database counted 92 hedge funds holding DDOG in Q2 2026, up from 80 in Q1; Arrowstreet Capital held 2,195,554 shares after increasing its stake 40%. DT holders rose to 48 from 46, while AQR Capital Management owned 8,317,467 shares after a 346% increase. Those quarter-end snapshots predate the September announcements.
As of August 31, 10,220,303 DDOG shares were sold short, equal to 3.04% of float and 2.94 days of average volume. That modest bearish position alone does not settle the thesis. The deciding question is whether autonomous agents prevent enough problems to reduce monitoring, or create enough software activity to make observability a larger machine-to-machine market.
READ NEXT: The Trillion-Dollar AI Capex Gap: Why Too Much Hardware Could Be Nvidia’s Trap and Microsoft’s Opportunity and Redditors Are Skeptical of SpaceX’s Orbital Data Centers. Are They Right?
