Photo: Designed by vectorjuice / Freepik
Although 100 percent of the surveyed organizations are already using AI in their operations, trust remains a challenge. Concerns about transparency, data quality, and governance often hinder AI projects in the transition from proof-of-concept to production.
The most common AI use cases are in data management (57%), AI governance (50%), and security (46%), but areas such as sustainability (27%) and log management (29%) are also rapidly gaining importance.
According to the report, business leaders see particular potential in AI-driven anomaly detection (41%) and real-time response to security risks (37%). These applications depend on a strong observability layer that brings together data from various systems and adds context.
Governance and security central to AI transition
One in four business leaders cites improving AI governance and trust as the top priority for the coming year. In decisions supported by AI, humans still play an important role: in 69 percent of cases, employees manually verify the accuracy of AI outcomes.
Additionally, 98 percent of organizations use AI to monitor compliance with security regulations. 69 percent saw the budget for AI-driven threat detection increase and expect further growth. The emphasis is shifting from reactive security to proactive risk management.
Observability as the foundation for AI-native IT
'Enterprise IT must evolve from adding AI to existing systems to building true AI-native experiences,' says Alois Reitbauer, Chief Technology Strategist at Dynatrace. 'This requires an observability layer that provides insight, context, and feedback loops — so that AI systems not only function smartly but also reliably and responsibly.'
According to the report, more than half of business leaders see automated real-time observability as key to a better customer experience within one year. Furthermore, 46 percent expect that the largest ROI from AI-driven observability will come from optimizing AI model configurations.
State of Observability Report 2025 looks ahead
By 2030, the majority of organizations expect to use AI for data encryption, risk assessment, and automated threat detection. Observability is shifting from a reactive measurement tool to a strategic decision-making system that provides insights to guide AI rather than just follow it.
Reitbauer concludes: 'Observability is transforming from telemetry to intelligence. As AI plays a larger role in business decisions, observability becomes the key to responsible, efficient, and scalable automation.'
