AI Inventory Intelligence Agent
AI Inventory Intelligence Agent automatically analyzes the AI asset inventory, identifies assets with incomplete metadata, and generates enrichment recommendations to improve data quality and governance readiness.
When the AI Inventory Intelligence Agent is active, it scans the inventory on a scheduled basis and evaluates each asset's metadata completeness. It presents actionable recommendations directly on the asset record. AI Stewards can accept a recommendation to update the asset automatically or dismiss it if the suggestion isn't applicable.
Inventory data completeness score
The inventory data completeness score measures how completely each managed AI asset's critical metadata fields are populated. The score is calculated as the percentage of required fields that contain values.
A higher completeness score indicates that an asset has sufficient metadata to support governance, risk assessment, and lifecycle management workflows. Assets with low completeness scores are prioritized in enrichment recommendations.
The completeness score appears on the AI asset inventory list and on individual asset records, giving AI Stewards a quick view of which assets require attention.
Note: The completeness score formula definition is available as a tooltip on the score field. The formula counts the number of critical fields populated out of the total required fields for that asset type.
Enrichment recommendations
Enrichment recommendations are field-level suggestions generated by the AI Inventory Intelligence Agent for assets with incomplete or missing metadata. Recommendations are sourced from two strategies:
- Deterministic enrichment: Rules-based suggestions derived from existing asset metadata such as vendor, model name, manufacturer, and category.
- AI-powered enrichment: LLM-generated suggestions for fields such as description, use cases, and documentation. AI-generated fields are marked with an AI indicator on the recommendation.
Each recommendation includes a confidence score and rationale, so AI Stewards can evaluate the suggestion before accepting it. Accepting a recommendation updates the asset field directly. Dismissing a recommendation leaves the field unchanged.
How the AI Inventory Intelligence Agent works
The agent operates as follows:
- The agent scans managed AI assets in the inventory on a scheduled basis.
- For each asset, it evaluates which critical metadata fields are missing or incomplete.
- It generates field-level enrichment recommendations using deterministic rules and LLM inference.
- Recommendations appear on the Enrichment Recommendations tab on the asset record.
- The inventory data completeness score is calculated and displayed for the asset.
- AI Stewards review recommendations and select accept (asset updated) or dismiss (no change made) for each suggestion.
- Accepted recommendations update the asset field; dismissed recommendations leave the field unchanged.
- The completeness score updates to reflect accepted changes.
Enabling the AI Inventory Intelligence Agent
To add LLM-generated suggestions for fields such as description, use cases, and documentation, turn on AI-generated enrichment. AI-generated enrichment is off by default, so the agent uses only deterministic enrichment until an AI Steward turns it on. Agent configuration settings, including model provider and skill activation, are managed in ServiceNow Otto admin.
Note: To use the AI Inventory Intelligence Agent, install the ServiceNow Otto for AI Control Tower (com.sn_aict_genai) plugin.
To turn on AI-generated enrichment, navigate to Settings > Rules and templates > AI skills and set the AI-generated enrichment toggle to Active. The toggle sets the sn_ai_governance.enrichment.generative_enabled system property to true, so enrichment runs both deterministic and generative (LLM-based) strategies. To use only deterministic enrichment, set the toggle to Inactive.
Only users with the AI Steward role can change this setting.
AI-generated enrichments are available with selected ServiceNow product tiers. For more information, see ServiceNow product tiers.
--- title: AI Inventory Intelligence Agent description: AI Inventory Intelligence Agent automatically analyzes the AI asset inventory, identifies assets with incomplete metadata, and generates enrichment recommendations to improve data quality and governance readiness. locale: en-US canonical_url: https://www.servicenow.com/docs/r/intelligent-experiences/inventory-intelligence-agent.html release: brazil topic_type: concept last_updated: "2026-09-10" reading_time_minutes: 2 breadcrumb: [Managing your AI asset inventory, Discover and manage AI assets, AI Control Tower, Establishing AI governance, Enable AI Experiences] --- # AI Inventory Intelligence Agent AI Inventory Intelligence Agent automatically analyzes the AI asset inventory, identifies assets with incomplete metadata, and generates enrichment recommendations to improve data quality and governance readiness. When the AI Inventory Intelligence Agent is active, it scans the inventory on a scheduled basis and evaluates each asset's metadata completeness. It presents actionable recommendations directly on the asset record. AI Stewards can accept a recommendation to update the asset automatically or dismiss it if the suggestion isn't applicable. ## Inventory data completeness score The inventory data completeness score measures how completely each managed AI asset's critical metadata fields are populated. The score is calculated as the percentage of required fields that contain values. A higher completeness score indicates that an asset has sufficient metadata to support governance, risk assessment, and lifecycle management workflows. Assets with low completeness scores are prioritized in enrichment recommendations. The completeness score appears on the AI asset inventory list and on individual asset records, giving AI Stewards a quick view of which assets require attention. **Note:** The completeness score formula definition is available as a tooltip on the score field. The formula counts the number of critical fields populated out of the total required fields for that asset type. ## Enrichment recommendations Enrichment recommendations are field-level suggestions generated by the AI Inventory Intelligence Agent for assets with incomplete or missing metadata. Recommendations are sourced from two strategies: - Deterministic enrichment: Rules-based suggestions derived from existing asset metadata such as vendor, model name, manufacturer, and category. - AI-powered enrichment: LLM-generated suggestions for fields such as description, use cases, and documentation. AI-generated fields are marked with an AI indicator on the recommendation. Each recommendation includes a confidence score and rationale, so AI Stewards can evaluate the suggestion before accepting it. Accepting a recommendation updates the asset field directly. Dismissing a recommendation leaves the field unchanged. ## How the AI Inventory Intelligence Agent works The agent operates as follows: - The agent scans managed AI assets in the inventory on a scheduled basis. - For each asset, it evaluates which critical metadata fields are missing or incomplete. - It generates field-level enrichment recommendations using deterministic rules and LLM inference. - Recommendations appear on the Enrichment Recommendations tab on the asset record. - The inventory data completeness score is calculated and displayed for the asset. - AI Stewards review recommendations and select accept \(asset updated\) or dismiss \(no change made\) for each suggestion. - Accepted recommendations update the asset field; dismissed recommendations leave the field unchanged. - The completeness score updates to reflect accepted changes. ## Enabling the AI Inventory Intelligence Agent To add LLM-generated suggestions for fields such as description, use cases, and documentation, turn on AI-generated enrichment. AI-generated enrichment is off by default, so the agent uses only deterministic enrichment until an AI Steward turns it on. Agent configuration settings, including model provider and skill activation, are managed in ServiceNow Otto admin. **Note:** To use the AI Inventory Intelligence Agent, install the ServiceNow Otto for AI Control Tower \(com.sn\_aict\_genai\) plugin. To turn on AI-generated enrichment, navigate to **Settings** > **Rules and templates** > **AI skills** and set the **AI-generated enrichment** toggle to **Active**. The toggle sets the sn\_ai\_governance.enrichment.generative\_enabled system property to true, so enrichment runs both deterministic and generative \(LLM-based\) strategies. To use only deterministic enrichment, set the toggle to **Inactive**. Only users with the AI Steward role can change this setting. AI-generated enrichments are available with selected ServiceNow product tiers. For more information, see [ServiceNow product tiers](https://www.servicenow.com/docs/r/intelligent-experiences/ai-native-sku-overview.html).
| title | AI Inventory Intelligence Agent |
| description | AI Inventory Intelligence Agent automatically analyzes the AI asset inventory, identifies assets with incomplete metadata, and generates enrichment recommendations to improve data quality and governance readiness. |
| locale | en-US |
| canonical_url | https://www.servicenow.com/docs/r/intelligent-experiences/inventory-intelligence-agent.html |
| release | brazil |
| topic_type | concept |
| last_updated | 2026-09-10 |
| reading_time_minutes | 2 |
| breadcrumb | Managing your AI asset inventoryDiscover and manage AI assetsAI Control TowerEstablishing AI governanceEnable AI Experiences |