Certified AI Records and Information Management Professional

What you will be able to do

• Use AI to classify content against a file plan and retention schedule
• Design automated metadata extraction with quality controls
• Apply retention and disposition with defensible human oversight
• Assess a repository's readiness to be used by AI tools
• Plan AI-assisted migration and clean-up of legacy content
• Keep records authentic, reliable and auditable when AI touches them

Modules

1. 1Records principles in an AI setting

   Authenticity, reliability, integrity and usability are the tests a record must pass. AI tools must support them, not wear them away.

   • Records versus documents versus data
   • The lifecycle from capture to disposition
   • Evidential value and chain of custody
   • Where AI helps and where it creates risk

   Exercise. List the points in your organization's records lifecycle where AI could act, and the integrity control each would need.

2. 2Auto-classification and the file plan

   Training and rules-plus-model approaches that map content to a business classification scheme.

   • Business classification schemes and functional file plans
   • Training sets drawn from well-filed content
   • Confidence thresholds and review queues
   • Measuring classification accuracy over time

   Exercise. Choose 200 documents to form a training and test set for one branch of a file plan, explaining your sample.

3. 3Metadata extraction and enrichment

   Pulling dates, parties, identifiers and subjects from documents, and controlling the quality of what is written back.

   • Intelligent document processing
   • Controlled vocabularies and authority lists
   • Validating extracted values
   • Provenance: recording that a value was machine-generated

   Exercise. Specify the metadata schema and validation rules for AI extraction from a set of contracts.

4. 4Retention, disposition and legal hold

   AI can propose retention classes and find content due for disposal. Decisions to destroy remain accountable to people.

   • Linking classifications to retention rules
   • Disposition review with AI-assisted sampling
   • Legal hold identification
   • Certificates of destruction and audit

   Exercise. Design a disposition approval workflow where AI proposes and a records officer confirms.

5. 5Repository readiness for AI

   Generative AI and search tools are only as good as the repository they read. Duplicates, stale versions and broken permissions all surface in answers.

   • Redundant, obsolete and trivial content
   • Version and duplicate control
   • Permission hygiene and over-sharing
   • Readiness scoring and remediation plans

   Exercise. Run a readiness assessment on one shared drive or library and estimate the share of content that is redundant, obsolete or trivial.

6. 6AI-assisted migration and clean-up

   Moving content between systems with AI handling classification, de-duplication and mapping, and people handling exceptions.

   • Content inventory and analysis
   • Mapping old structures to new
   • De-duplication and near-duplicate detection
   • Migration validation and sign-off

   Exercise. Write the validation steps that prove a migrated collection is complete and its metadata is intact.