Datadog Enhances AI Security and Observability Capabilities
Severity: Low (Score: 39.9)
Sources: Pulse2, www.datadoghq.com, Heise.De, docs.datadoghq.com
Published: · Updated:
Keywords: datadog, observability, platform, security, autonomous, capabilities, annual
Severity indicators: pla
Summary
At the DASH conference on June 9-10, 2026, Datadog introduced over 100 new features aimed at enhancing its AI observability and security platform. Key announcements included the expansion of the Bits AI suite, which now supports autonomous detection, investigation, and remediation of issues across various environments. The new AI Guard tool is designed to protect against attacks targeting AI agents, utilizing telemetry tracing and behavioral analysis. Additionally, the Bring Your Own Cloud Logs feature allows customers to manage log data in their own storage systems. These innovations aim to address the operational complexities arising from rapid AI adoption and software development. The enhancements also include new capabilities for debugging AI agents and integrations with collaboration tools like Slack. Datadog emphasizes the importance of maintaining operational control in increasingly complex environments. Key Points: • Datadog launched over 100 new capabilities at the DASH conference to enhance AI operations. • The AI Guard tool aims to detect and block sophisticated attacks on AI agents. • New features support autonomous workflows and allow customers to manage log data in their own cloud.
Detailed Analysis
**Impact** Enterprises adopting AI-driven observability and security platforms are affected, particularly those in sectors with high operational complexity such as finance and software development. The new capabilities target organizations managing rapid AI integration and large-scale telemetry data, with a focus on DACH-region companies requiring compliance with ITIL and ISO 27001 standards. Risks include exposure to sophisticated AI agent attacks like prompt injection and agent poisoning, potentially leading to sensitive data leaks and operational disruptions. The enhancements aim to reduce incident resolution times and support autonomous remediation within controlled guardrails. **Technical Details** Attack vectors include multi-stage prompt injection and malicious prompt attacks targeting AI agents with elevated privileges. Detection uses a combination of telemetry tracing and AI-native stateful behavioral analysis to identify anomalies missed by stateless prompt-response systems. The Bits AI SRE agent autonomously investigates root causes by querying metrics, logs, traces, and topology data, integrating with tools such as Confluence, ServiceNow, and Jira. No specific malware, CVEs, or IOCs are disclosed in the provided information. **Recommended Response** Deploy AI Guard or equivalent telemetry and behavioral analysis tools to detect and block prompt injection and agent poisoning attacks. Implement strict RBAC, policy enforcement, audit logging, and require human approval for automated remediation in security-critical environments. Integrate AI agents within existing ITIL or ISO 27001 compliant workflows and monitor AI agent activity via centralized consoles. Maintain vigilance for anomalous agent behavior and ensure secure configuration of AI agent deployment and custom automation workflows.
Source articles (4)
- Datadog: 100+ New Capabilities Introduced To Advance Autonomous Operations And AI Security — Pulse2 · 2026-06-09
Datadog unveiled more than 100 new products and capabilities at its annual DASH conference, expanding its AI-powered observability and security platform to help organizations manage growing operationa… - Datadog expands observability platform to autonomous AI team colleague — Heise.De · 2026-06-10
At its annual DASH conference, Datadog announced numerous new features for its observability and security platform. The focus is on the AI agent Bits AI SRE, the AI Guard tool for protecting AI applic… - Ai Guard — www.datadoghq.com · 2026-06-10
Organizations are increasingly using agentic AI applications powered by large language models (LLMs) to automate analysis, decision-making, and operational workflows. As these AI agents take on more r… - Sdk — docs.datadoghq.com · 2026-06-10
Use an SDK to call the AI Guard REST API and monitor AI Guard activity in real time in Datadog. SDKs use the Datadog Agent to send AI Guard data to Datadog. The Agent must be running and accessible to…
Timeline
- 2026-06-09 — DASH conference announcements: Datadog unveiled over 100 new features for its AI observability and security platform, focusing on operational complexity and AI adoption.
- 2026-06-10 — Bits AI expansion announced: Datadog expanded its Bits AI suite to include autonomous detection and remediation capabilities.
- 2026-06-10 — AI Guard tool introduced: Datadog launched AI Guard to protect against attacks on AI agents, utilizing advanced telemetry and behavioral analysis.
Related entities
- Data Breach (Attack Type)
- Germany (Country)
- CWE-200 - Exposure of Sensitive Information (Cwe)
- german.it (Domain)
- infrastructure.ai (Domain)
- T1041 - Exfiltration Over C2 Channel (Mitre Attack)
- T1567 - Exfiltration Over Web Service (Mitre Attack)
- Curl (Tool)