Research Finder
Find by Keyword
Dynatrace Doubles Down on AI: Observability is the Key
Dynatrace expands its observability platform with new features designed to support generative AI initiatives.
Key Takeaways
- Dynatrace introduces enhanced AI observability capabilities.
- The platform now offers features such as LLM Model Analytics, LLM Input and Output Guardrails, Multi-model Tracing, and Responsible AI Integrations.
- These advancements aim to help organizations better manage their AI applications, ensuring reliability, performance, security, and compliance.
- AI observability is crucial for organizations looking to ensure their AI applications work as intended and maximize the value of their AI investments.
The News:
On January 28, 2025 Dynatrace announced new AI observability features designed to provide comprehensive insights into AI applications. These capabilities aim to deliver greater visibility into AI performance, cost, and compliance. Dynatrace's new AI observability features include:
- Enhanced LLM Model Analytics: This feature provides detailed insights into the performance and cost of large language models (LLMs). It tracks key metrics such as input and output errors, response times, and token consumption. Additionally, it leverages Dynatrace Davis AI® to predict cost changes associated with LLM usage.
- LLM Input and Output Guardrails: Dynatrace has introduced guardrails designed to ensure the quality and safety of AI application input and output. These guardrails help identify and prevent issues such as model hallucinations, malicious prompt injection, PII leakage, and toxic language.
- Multi-model Tracing: This capability enables organizations to trace dependencies between multiple LLMs working together in Retrieval Augmented Generation (RAG) pipelines or agentic frameworks. It provides end-to-end visibility into the entire AI system, allowing teams to identify and resolve performance bottlenecks and ensure optimal user experiences.
- Responsible AI Integrations: Dynatrace now offers integrations designed to support responsible AI development and deployment. These integrations help organizations track every input and output, providing an audit trail for monitoring and observability. They also document the training data used for a given model, promoting transparency and accountability. Dynatrace Grail™ allows organizations to query this data on demand.
Analyst Take:
The rapid rise of generative AI has created a critical need for robust AI observability solutions. Organizations are increasingly adopting AI-powered applications, and they need tools to help them understand and manage these complex systems. Dynatrace's latest advancements in AI observability are well-timed and aim to address this growing demand.
The advancements announced by Dynatrace are significant because they address several key challenges associated with AI observability. First, they provide comprehensive visibility into the performance of AI models. This is critical for ensuring that AI applications are functioning as expected and delivering the desired outcomes. Second, the new capabilities help organizations manage the cost of AI. AI can be expensive to operate, and Dynatrace's tools aim to provide insights into how to optimize AI usage and reduce costs. Third, Dynatrace's advancements address the growing need for responsible AI. As AI becomes more prevalent, it is essential to ensure that it is being used ethically and responsibly.
Another significant aspect of Dynatrace's announcement is the focus on multi-model tracing. AI applications are increasingly relying on multiple models working together. Dynatrace's ability to trace dependencies between these models provides valuable insights into how they are interacting and can help identify potential bottlenecks or issues.
It’s notable that Dynatrace chose to announce these new features a week before their annual Perform conference. While we may have expected these advancements to be unveiled at the event itself, this pre-conference announcement strategically builds anticipation and allows for a deeper dive into the capabilities during the conference sessions. This approach not only generates excitement for Perform 2025 but also demonstrates Dynatrace's confidence in their AI observability solutions.
Overall, Dynatrace's latest advancements in AI observability represent a significant step forward. They provide organizations with the tools and insights they need to effectively manage their AI applications and achieve their desired outcomes. As AI continues to evolve and become more sophisticated, the importance of AI observability will only grow. Dynatrace is well-positioned in this emerging market.
Looking Ahead
AI observability is essential for organizations looking to ensure their AI applications are working correctly and maximize the value of their IT investments. Without proper observability, organizations can’t understand how applications and infrastructure are performing, identify potential issues, and detect anomalies early before they become an outage. AI observability capabilities will only become more important as organizations try to effectively manage their AI applications and achieve their desired outcomes.
Based on our observations, Dynatrace's move to enhance its AI observability capabilities is a strategic and timely decision. The market for AI observability solutions is expected to grow substantially in the coming years, and Dynatrace is well-positioned to capitalize on this trend. My perspective is that Dynatrace is a company to watch in the AI observability space and going forward we will be monitoring how the company performs on delivering its promises and meeting the evolving needs of its customers. When you look at the market as a whole, the announcement today positions Dynatrace very favorably in the AI observability space.
Stephanie Walter | Analyst In Residence - AI Tech Stack
Stephanie Walter is a results-driven technology executive and analyst in residence with over 20 years leading innovation in Cloud, SaaS, Middleware, Data, and AI. She has guided product life cycles from concept to go-to-market in both senior roles at IBM and fractional executive capacities, blending engineering expertise with business strategy and market insights. From software engineering and architecture to executive product management, Stephanie has driven large-scale transformations, developed technical talent, and solved complex challenges across startup, growth-stage, and enterprise environments.
Steven Dickens | CEO HyperFRAME Research
Regarded as a luminary at the intersection of technology and business transformation, Steven Dickens is the CEO and Principal Analyst at HyperFRAME Research.
Ranked consistently among the Top 10 Analysts by AR Insights and a contributor to Forbes, Steven's expert perspectives are sought after by tier one media outlets such as The Wall Street Journal and CNBC, and he is a regular on TV networks including the Schwab Network and Bloomberg.