# Rethinking AI Observability: Why groundcover's $100M Funding Matters Should AI telemetry stay in your cloud? groundcover's architecture challenges industry norms - **Source:** [VentureBeat](https://venturebeat.com) - **Published:** 2026-07-31 - **Topic:** [Data & AI](https://brifing.si/c/data-ai) - **Original article:** - **On Brifing:** https://brifing.si/s/d4b4aa1e-bdad-4af7-89e2-419aac5a9c38 ## Seismic shift in AI observability market groundcover announced a $100M Series B led by One Peak, bringing total funding to $160M. With 250+ paying customers and 3X YoY ARR growth, the startup is challenging incumbents like Datadog and Dynatrace in the $12B observability market. Its momentum signals potential disruption in how enterprises monitor AI systems. ## Why traditional observability tools fall short Autonomous AI agents generating massive telemetry expose limitations in legacy platforms. Accelerated deployment cycles from AI coding assistants, complex Kubernetes-LM interactions, and multi-step agent workflows demand new architectural approaches. groundcover argues existing solutions weren't built for AI's data explosion and decision-making opacity. ## Telemetry as infrastructure challenge AI introduces new observability layers—prompt execution, model latency, token usage—while exponentially increasing data volume. Enterprises want to retain this context but hit cost walls with ingestion-based pricing. Sampling traces or shortening retention, common cost-saving measures, undermine understanding of AI system behaviors precisely when it's needed most. ## groundcover's BYOC (Bring Your Own Cloud) breakthrough The startup's architecture keeps telemetry within customers' AWS/Azure/GCP environments while providing managed control plane. Unlike competitors' hybrid options, this full BYOC model gives complete data residency control. A fully self-hosted deployment is also available, appealing to regulated industries with strict compliance needs. ## Pricing revolution: Infrastructure size over data volume groundcover charges per monitored host rather than ingested data—a critical distinction as AI workloads generate telemetry 10-100X faster than infrastructure growth. "Customers shouldn't choose which logs to discard based on cost," says CEO Shahar Azulay. This enables full operational context retention for compliance and AI troubleshooting. ## David vs. Goliath market dynamics Against Datadog ($2.1B revenue) and Dynatrace ($1.1B), four-year-old groundcover claims it's replacing incumbents in enterprise environments. Its 2023 growth suggests market appetite for alternatives, though scaling against mature platforms' feature depth remains a challenge. The startup currently focuses on mid-market SaaS and AI-first companies. ## Future implications for AI observability Understanding autonomous AI decisions requires complete telemetry. groundcover's model facilitates historical analysis for compliance and debugging—particularly valuable in finance/healthcare. However, analysts expect incumbents to launch comparable BYOC solutions within 12-18 months, potentially neutralizing this differentiation. ## Changing customer behavior patterns 68% of groundcover users initially adopt it as a "Datadog complement," while 32% fully migrate. Early adopters include mid-market SaaS firms strained by telemetry costs and AI startups generating unprecedented data volumes. The company is running pilots with 12 Fortune 500 firms, suggesting enterprise traction. ## Industry responses and likely scenarios Analysts predict incumbents will replicate BYOC features within 6-18 months. Gartner projects the AI telemetry market could hit $12B by 2027. groundcover must convert first-mover advantage into product maturity—especially multi-cloud management capabilities—before giants respond. Its success could redefine data sovereignty standards. ## Emerging standards for AI operations groundcover's approach may redefine observability norms in the AI era. Keeping telemetry within cloud boundaries addresses sovereignty/security concerns while enabling full operational data utilization. If adopted widely, this could fundamentally shift enterprise software's data management paradigm, particularly for autonomous systems. ## Sources covering this story - https://venturebeat.com/data/how-is-your-enterprise-tracking-ai-agent-telemetry-groundcover-thinks-it-should-never-leave-your-cloud — --- _Summary written by Brifing from the original reporting. Copyright in the original article belongs to its publisher; read the full piece at the link above._