In progressAI Agents
Human-in-the-loop governance
The navigation is ahead of the writing. This route is part of the planned information architecture and the page has not been built yet.
What will live here
Every AI result waits for a person. Nothing reaches your compliance record until someone approves it.
The gap is in this site, not in the platform. What CarcGRC does is specified in detail — eight AI agents that stage their work for a human decision, six frameworks at launch, a dedicated deployment per customer. We would rather leave a route honestly unfinished than fill it with copy we cannot trace back to the spec.
Route status
| Field | Value |
|---|---|
| route | /ai-agents/governance |
| status | in progress |
| section | AI Agents |
| product | pre-launch · Phase 1 |
Elsewhere in AI Agents
- Gap AgentReads every control and its evidence, then files a severity-ranked list of gaps with a suggested fix for each one.
- Evidence AgentRe-evaluates control health after every integration sync and quality-scores each collected evidence item from 1 to 10.
- Risk AgentTurns vulnerability and threat findings into scored risk register entries, and flags exceptions approaching expiry.
- Policy AgentDrafts required policies from the org’s stack context and flags configuration drift that conflicts with a published policy.