Delivering operational trust for AI agents

Share on:

AI agents are moving into production faster than organizations can govern them. And “operational trust” is more critical than ever.

According to Deloitte's 2026 State of AI in the Enterprise survey of 3,235 business and IT leaders across 24 countries, 74% of organizations expect to use AI agents at least moderately by 2027, up from 23% today.

Yet only 21% have a mature governance model in place for agentic AI; in other words, roughly 80% lack capabilities like clear agent decision boundaries, real-time monitoring and audit trails.

Closing the gap between agent adoption and agent trust requires trust you can measure continuously. Not assess once.1

At Collibra, we’re expanding capabilities for the Collibra AI Command Center so our customers can continuously measure operational trust across every AI agent, using evaluation signals generated throughout the AI lifecycle. From pre-production validation to production runtime.

What's new: Operational trust for AI agents

Generally available alongside new integrations with Databricks Agent Bricks, Microsoft Foundry, Amazon Bedrock and Google Gemini, our newest capability transforms LLM-as-a-judge evaluation results into a dynamic AI Trust Score for every deployed agent wherever it runs, while surfacing additional operational insights such as AIUC-1 certification status and token consumption trends.

The new Agents dashboard rolls these signals up across your entire AI fleet, regardless of which platform each agent is built on.

At a glance, you can see:

How many agents are running
The average Trust Score across the fleet
Your evaluation coverage gap
The pass rate across all AI monitors

Plus, you’ll see a 30-day rolling trend that makes degradation visible the moment it starts.

With this update, trust is now a living measure that evolves with each agent's real-world behavior.

How operational trust for AI agents helps

Most organizations approve AI agents once — at deployment — and then lose sight of them.

Runtime behavior drifts as models are updated, prompts change and usage grows, but governance teams have no continuous signal telling them whether an agent still deserves the trust it was granted. Evaluations run ad hoc, monitoring sits in engineering tools that governance can't see, and by the time quality or safety issues surface they've already reached users.

As agent fleets grow into the hundreds or thousands (and spread across multiple agent platforms) manual spot-checking simply can't keep up.

How Operational Trust for Agents closes the gap

Trust becomes outdated after deployment. Operational Trust continuously assesses how agents behave in production, so trust reflects what is happening now, not just what was approved at launch.
Teams lack a common measure of trust. A shared AI Trust Score gives governance, engineering, and business teams an objective view of agent trustworthiness.
Quality and safety issues surface too late. Continuous runtime signals help teams detect degradation and drift before they become business-impacting incidents.
Evaluation coverage has blind spots. Fleet-level visibility shows which agents are adequately evaluated — and where evaluation gaps remain.
Runtime monitoring is fragmented across platforms. Operational Trust brings signals from Databricks, Microsoft Foundry, Amazon Bedrock, Google Gemini, and other platforms into one governance view connected to the agent registry.

How operational trust for AI agents works

The AI Trust Score is a dynamic 0–100 rating computed for every deployed agent registered in the AI Command Center.

Rather than relying on documentation or one-off assessments, the score is derived from LLM-as-a-judge evaluation results that continuously assess key aspects of agent behavior, such as groundedness, faithfulness, relevance, correctness, retrieval quality and safety, across pre-production and production.

AIUC-1 compliance status and token consumption trends are surfaced alongside the Trust Score as complementary operational signals, providing a more complete view of each agent's operational health.

As new evaluation results arrive, each agent's Trust Score updates continuously. For example, a score of 27/100 tells you immediately that trust has eroded and why.

*Track the health of your AI agent ecosystem through unified KPIs, AI Trust Scores, evaluation coverage, and operational trends — all from a single dashboard.*

With the new agent platform integrations (GA, available now), agents built and served in Databricks Agent Bricks, Microsoft Foundry, Amazon Bedrock and Google Gemini are connected directly to the AI Command Center. Their runtime telemetry (including evaluations, judge verdicts and usage metrics) flows into Collibra automatically, so agents running on any of these platforms are registered, scored and monitored alongside the rest of your AI estate, without custom pipelines. Because every agent is scored on the same scale, you get one consistent view of trust across a heterogeneous fleet instead of four separate dashboards.

Everything rolls up in the Agents dashboard inside the AI Command Center. KPI tiles summarize the fleet, including:

Total AI agents
Average Trust Score
Evaluation coverage gap (agents lacking sufficient evals)
Mean pass rate across all AI monitors

A rollup heatmap distributes agents across configurable High, Medium and Low trust bands by lifecycle stage (including from candidate through approval) so you can spot concentrations of low-trust agents instantly. A 30-day rolling chart of the average AI Trust Score reveals fleet-wide trends in degradation before they become incidents.

*Continuously monitor the health of your AI agents with real-time quality, compliance, and consumption metrics — turning operational signals into enterprise-wide AI trust.*

Running on Collibra Platform, every score is anchored to the governed agent record in the AI registry: its owner, use case, policies, underlying models and connected data assets.

Now, trust isn't just a metric. With the AI Command Center, trust is an operational signal that can trigger governance workflows, review tasks and status changes in the same platform where the agent was approved. And it could mean the difference between the success and failure of your agentic efforts.

1 Deloitte, State of AI in the Enterprise, Jan 2026