Multi-Instance Framework for AI value data
The Multi-Instance Framework (MIF) synchronizes AI value data from a sub-production instance to a production instance only for Creator skills. In this setup, AI Control Tower calculates value for Creator skills using data from development and testing environments.
For creator skills deployed in a sub-production instance, you can evaluate how the value is determined for an AI system by running a metric within its value template. The results are synced to the production instance for the value calculation.
To facilitate this process, MIF connects the sub-production instance to the production instance, which functions as the manager instance. MIF then synchronizes value data for Creator skills to the production instance.
Note:
MIF supports value calculation on sub-production instances only for Creator skills. Value calculation on sub-production instances isn't available for any other skills.
Value data that MIF syncs
MIF syncs the following value data from the sub-production instance to the production instance:
- Creator metrics data for all creator skills
- Results from testing value templates for AI systems in the sub-production environment
Related topics
--- title: Multi-Instance Framework for AI value data description: The Multi-Instance Framework \(MIF\) synchronizes AI value data from a sub-production instance to a production instance only for Creator skills. In this setup, AI Control Tower calculates value for Creator skills using data from development and testing environments. locale: en-US canonical_url: https://www.servicenow.com/docs/r/intelligent-experiences/mv-multi-instance-framework-for-ai-value-data.html release: brazil topic_type: concept last_updated: "2026-09-25" reading_time_minutes: 1 keywords: [multi-instance framework, MIF, AI value, sub-production instance] breadcrumb: [Value, Explore, Measure AI system, Measure AI systems, AI Control Tower, Establishing AI governance, Enable AI Experiences] --- # Multi-Instance Framework for AI value data The Multi-Instance Framework \(MIF\) synchronizes AI value data from a sub-production instance to a production instance only for Creator skills. In this setup, AI Control Tower calculates value for Creator skills using data from development and testing environments. For creator skills deployed in a sub-production instance, you can evaluate how the value is determined for an AI system by running a metric within its value template. The results are synced to the production instance for the value calculation. To facilitate this process, MIF connects the sub-production instance to the production instance, which functions as the manager instance. MIF then synchronizes value data for Creator skills to the production instance. **Note:** MIF supports value calculation on sub-production instances only for Creator skills. Value calculation on sub-production instances isn't available for any other skills. ## Value data that MIF syncs MIF syncs the following value data from the sub-production instance to the production instance: - Creator metrics data for all creator skills - Results from testing value templates for AI systems in the sub-production environment **Related topics** [Set up Multi-Instance Framework for value calculations](https://raw.githubusercontent.com/ServiceNow/ServiceNowDocs/brazil/markdown/intelligent-experiences/mv-set-up-the-multi-instance-framework-for-value-calculations.md)
| title | Multi-Instance Framework for AI value data |
| description | The Multi-Instance Framework \(MIF\) synchronizes AI value data from a sub-production instance to a production instance only for Creator skills. In this setup, AI Control Tower calculates value for Creator skills using data from development and testing environments. |
| locale | en-US |
| canonical_url | https://www.servicenow.com/docs/r/intelligent-experiences/mv-multi-instance-framework-for-ai-value-data.html |
| release | brazil |
| topic_type | concept |
| last_updated | 2026-09-25 |
| reading_time_minutes | 1 |
| keywords | multi-instance frameworkMIFAI valuesub-production instance |
| breadcrumb | ValueExploreMeasure AI systemMeasure AI systemsAI Control TowerEstablishing AI governanceEnable AI Experiences |