When AI agents become operational in security, you will need them not only to be monitoring and responding to security incidents, but you also need to be able to monitor and respond to agent based decisions. This is where we need to shift and tie observability of our agents directly with our security capability. Traditionally, we as security professionals have been focused on actions occurred; now, we must switch to why they were performed:

What evidence did the agent consider?
What hypotheses did it explore and discard?
What reasoning led to action?
Did it hallucinate a connection between two unrelated events?
Did it skip a critical investigation step?
Did its confidence score reflect genuine signals or did it create one?

This semantic gap is telemetry and assessment of that telemetry. One customer that has built a system to assess agent behavior across its estate is Uber with its Agentic Detection and Response (ADR) system. The team so far has deployed this across 7,000+ hosts, where they are looking at the typical signals we look for in security like file writes and API calls, but they have extended this to include agent reasoning, prompts, and chains of intent leading to execution.

For security agents, specifically, the telemetry requirements are different from software engineering based agents, but the principle is identical since we want to capture:

The full reasoning chain: What alert data it ingested, what knowledge base entries it retrieved, what hypotheses it generated, how it weighted evidence, and what led to the classification decision.
The evidence: Every claim the agent makes should be traceable to a source.
The decision boundaries: Where did the agent’s reasoning change a decision? What alternative paths did it consider? When it chose to escalate rather than close, what was the evidence that tipped the decision?
The true negatives: What the agent didn’t do is just as important as what it did.

Uber's ADR system demonstrates a way of doing this at enterprise scale, pairing high-fidelity agentic telemetry capture with two-tier detection. The Sensor forwards reconstructed agent sessions to a backend for analysis and points to Elasticsearch as the telemetry store underpinning it. The paper validates that closing the telemetry gap at enterprise scale is both achievable and necessary.

Context engineering

Vector database

Search powered applications

Logs

Threat protection

Workflows

Elasticsearch

Kibana (Discover, Dashboards)

Elastic Agent Builder

AutoOps

Piped query language

Jina AI search models

Elastic Cloud Serverless

Elastic Cloud Hosted

Self-managed Elasticsearch

Ecommerce search

Customer support search

Search-driven apps

Log analytics

Infrastructure monitoring

Digital experience monitoring

App performance monitoring

AIOps

LLM observability

Next-gen SIEM

Workflows for security

XDR and endpoint security

AI for security

10x your data's value

Cloud providers

Elastic AI Ecosystem

Search AI Partner Program

AV-Comparatives

Forrester Wave™ XDR

Gartner Magic Quadrant Leader

IDC MarketScape

Search

Security

Observability

Get started

Demo gallery

Downloads

Integrations

Docs

Elasticsearch Labs

Elastic Security Labs

Elastic Observability Labs

Blog

Community

Events

Webinars

Discuss

Training

Support

Consulting

Author

Articles by Marcus Jeffes

Solutions Architect

The future of governing AI agents

Sign up for Elastic Cloud free trial