Datadog unveils new autonomous ops and AI-agent security tools

Datadog has unveiled more than 100 new capabilities aimed at helping organisations run operations more autonomously and secure the AI agents now embedded across their systems.

The centrepiece is an expansion of Bits AI, Datadog’s suite of agents for development, security and operations work.

With a set of new modules, Datadog said Bits AI can now autonomously detect, investigate and remediate infrastructure issues around the clock within pre-defined guardrails, recommending and applying fixes rather than only diagnosing root causes.

New evaluation features also let it debug and generate fixes for other AI agents, and it follows each code change and pull request from staging through to production. Bits AI is available in tools teams already use, including Slack and Claude.

On security, a new product called AI Guard is designed to protect AI agents from attacks such as prompt injection and agent poisoning.

Datadog said it combines deep agent telemetry tracing with behavioural anomaly analysis to catch attacks that stateless prompt-and-response checks miss - the risk being that a single hidden instruction can turn a well-intentioned agent, which may hold elevated privileges and access sensitive data, into a leak.

Datadog also addressed the cost of rising log volumes with Bring Your Own Cloud, which deploys its platform into a customer’s own environment so data is processed and indexed in the customer’s cloud object storage.

A pair of agent-management tools - Bits Agent Builder and Agent Console - were also introduced, letting teams build custom agents within their own controls and centrally monitor AI agents and coding tools such as Claude Code, Cursor and GitHub Copilot, correlating the work those agents produce with what they cost.

“We consistently invest about 30 per cent of revenue into R&D, which is why we are able to deeply understand and solve the problems our customers face every day in managing operational complexity,” said Alexis Lê-Quôc, co-founder and CTO at Datadog.

“At [our annual conference] DASH, we launched 100+ capabilities unified around one goal: giving customers the visibility they need to find and fix the issues that matter most, the moment they matter.”