Dynatrace has announced enhanced observability features specifically designed for customers utilizing Generative AI (GenAI) applications. These new capabilities offer comprehensive insights that aim to improve reliability, performance, security, and compliance in AI deployments, enabling organizations to better understand the returns on their investments.

As the demand for AI-supporting technologies surges, with worldwide spending expected to exceed $749 billion by 2028, Dynatrace’s advancements are positioned as essential for enterprises looking to capitalize on AI's potential. The company’s platform improvements include enhanced analytics for large language models (LLMs), which allow for monitoring and predicting usage behavior and associated costs, as well as mechanisms to ensure the quality and safety of AI application inputs and outputs.

Among the platform advancements are LLM input and output guardrails to mitigate issues like model hallucinations and PII leakage, and multi-model tracing which aids in observing entire systems rather than isolated components. Dynatrace's focus on responsible AI integrations allows for detailed tracking of AI-related data, ensuring compliance and aiding governance efforts.

Customer feedback highlights the value of these improvements, with industry leaders like FreedomPay citing enhancements in analytics and security that align with their compliance and operational goals. IDC’s Stephen Elliot emphasized the importance of observability in producing reliable and cost-effective AI solutions, underlining the need for real-time monitoring to detect anomalies and ensure data quality.

Alois Reitbauer, Chief Technology Strategist at Dynatrace, noted the growing trend of enterprises moving AI applications into production and stressed the significance of AI observability in achieving ROI and governance. The upcoming Perform 2025 conference will feature AI Observability prominently, showcasing Dynatrace’s ongoing commitments to supporting its customers’ AI ambitions.