AI Project Charter

Purpose

An AI Project Charter does two things a standard project charter does not: it locks in AI governance commitments before any build work begins, and it forces the organization to classify the system under applicable regulation at the point of authorization — not after deployment. It is the document that asks 'are we authorized to build this, and do we understand what that authorization commits us to?' before the question becomes 'why weren't we prepared for this?'

The sponsor's signature on this charter authorizes work and budget. It does not approve the AI system for deployment. Deployment authorization is a separate gate, requiring completion of all governance checkpoints listed in the charter. Conflating the two is one of the most common sources of premature deployment in AI projects.

Where It Fits in Your Document Pack

The charter establishes the governance foundation everything else builds on. The AI Impact Assessment references the system context and regulatory classification documented here. The AI Risk Register's scope is bounded by the in-scope and out-of-scope decisions made here. The stakeholder registry here determines who signs off on downstream governance documents.

What This Template Covers

• Project identification: charter ID, version, status, project manager, sponsor, and authorization date
• Business case and objectives: problem statement, expected benefits, measurable success criteria, and preliminary budget
• AI system details: AI approach, model source, intended use, and training and input data sources
• Governance and compliance: EU AI Act risk classification, NIST AI RMF alignment, regulatory requirements, and required governance checkpoints
• Scope definition: in scope, out of scope, key constraints, key assumptions, and dependencies
• Stakeholder registry: named individuals for sponsor, PM, AI lead, data owner, governance, and legal roles
• Budget and resources: authorized budget, FTE budget, and resource summary
• Key milestones: five milestone rows with target date, owner, and status — aligned to AI lifecycle stages
• Approvals and sign-off: three-signature block with sponsor, PM, and governance roles
• Revision history table and a completion guidance page with field-by-field instructions

Download