Public Records and AI

Knowledge Bundle: Public-Sector AI Readiness Concept ID: government:public_records Concept Type: Records and Transparency Guidance Status: Draft Audience: Public agencies, clerks, legal counsel, IT teams, department leaders Steward: OKF Expert AI Use: Human-reviewed Last Reviewed: June 2026

Public agencies should consider how AI-generated content, prompts, uploaded documents, chat histories, and AI-assisted work may intersect with public-records, retention, transparency, and disclosure obligations.

AI does not remove an agency’s responsibility to preserve records that are otherwise required to be retained.

• Whether prompts or AI-generated outputs qualify as agency records
• Whether communications created with AI must be retained
• Whether AI systems store chat history or uploaded files
• Whether vendor systems retain agency data
• Whether staff can export records when needed
• Whether AI-assisted documents require human review before release
• Whether the agency can explain how AI was used in a public-facing workflow

• Use approved AI tools only.
• Avoid entering confidential or protected information into public AI systems.
• Document agency-approved AI workflows.
• Maintain records according to applicable retention schedules.
• Preserve material AI-assisted work products when required.
• Establish clear rules for chat logs, prompts, uploaded files, and generated drafts.
• Consult records-management and legal staff before deploying public-facing AI systems.

• Does the AI platform retain prompts, uploads, or outputs?
• Can the agency export records on demand?
• Can records be deleted according to retention rules?
• Are audit logs available?
• Who owns the data submitted to the AI system?
• Does the vendor use agency data for model training?
• Can the agency respond to a public-records request involving AI-assisted work?

• AI Governance
• AI Risk Assessment
• AI Procurement
• Staff AI Training
• Return to Public-Sector AI Readiness