Datadog Unveils New AI Security Tools, Expands Cloud Monitoring

Datadog has introduced a suite of advanced AI agents and new security capabilities designed to help businesses manage and secure AI-driven applications more effectively. Unveiled at the DASH 2025 conference, these domain-specific AI agents aid in incident response, development, and security operations by automating tasks such as troubleshooting, alert management, and code fixes, thereby enhancing developer productivity and operational efficiency. The company also launched new security features, including Datadog Code Security, which detects and prioritizes vulnerabilities in custom code and open-source libraries, addressing risks from AI workloads and third-party code repositories. Datadog emphasizes the growing security challenges AI introduces due to its non-deterministic nature and expanded attack surfaces, with autonomous microservices able to perform actions without human oversight, potentially leading to cascading breaches. These innovations reflect Datadog's commitment to providing continuous runtime visibility and comprehensive protection across the AI development lifecycle, from coding to production environments. Additionally, Datadog is fostering transparency and innovation by releasing a new AI model for time series analysis as an open-weight resource on the Hugging Face platform.