# The Marshall AI Governance Standard Version: 0.2 Phase: Founder Phase Steward: Richard K. Marshall Website: https://marshall.net --- # Introduction The Marshall AI Governance Standard is a practical governance framework designed to help organizations adopt artificial intelligence tools responsibly while maintaining clear human accountability. Artificial intelligence capabilities are entering organizations rapidly through everyday productivity software, research tools, and automated systems. In many cases these tools appear in operational workflows before leadership establishes governance structures. The Marshall Standard provides a simple operational structure that allows organizations to introduce responsible oversight early in the adoption process. --- # The Marshall Principle Artificial intelligence may assist human decision-making. Responsibility always remains with humans. This principle defines the philosophical foundation of the Marshall AI Governance Standard. Artificial intelligence systems may provide analysis, suggestions, or generated outputs, but accountability for decisions must remain with identifiable human actors. --- # Governance Objectives The Marshall Standard focuses on five governance objectives. 1. Establish visibility into where artificial intelligence is being used. 2. Define boundaries for what information may be shared with AI systems. 3. Ensure human responsibility for decisions assisted by AI tools. 4. Maintain awareness of external AI service providers and associated risks. 5. Introduce governance practices before operational failures occur. --- # Governance Domains The Marshall Standard evaluates readiness across five governance domains. ## Governance Authority Organizations must clearly identify who holds responsibility for AI governance decisions. This may include executive leadership, technology leadership, or designated governance committees. The key requirement is that authority is explicit rather than assumed. --- ## AI Visibility Organizations must understand where AI tools are already being used. This includes: - generative AI assistants - automated analysis tools - AI-enabled software platforms - internal machine learning systems Visibility is the first step toward responsible governance. --- ## Data Governance Organizations must define boundaries for information shared with AI systems. Sensitive information such as confidential client data, financial information, proprietary research, or protected personal information must be governed by clear policies. --- ## Human Oversight Artificial intelligence may assist analysis or recommendations, but responsibility for decisions must remain human. Organizations should ensure that AI-generated outputs are reviewed and validated before operational decisions are made. --- ## Vendor Governance Many AI tools operate as external cloud services. Organizations should maintain awareness of: - vendor policies - data handling practices - system limitations - contractual responsibilities --- # Participation Organizations may identify themselves as participating in the Marshall AI Governance Standard if they are actively working to introduce governance practices aligned with the framework. Participation represents engagement with governance principles rather than certification of perfect compliance. --- # AI Governance Readiness Assessment Organizations typically begin their participation by running the AI Governance Readiness Assessment. The assessment is designed to provide a quick signal about governance readiness and identify areas where oversight may be missing. The assessment is intentionally short and designed for leadership teams rather than technical specialists. --- # Reporting Expectations Organizations participating in the Marshall Standard are encouraged to review their governance practices regularly. Quarterly reporting periods provide natural checkpoints for reviewing AI usage and governance practices. Suggested reporting dates: March 31 June 30 September 30 December 31 --- # Verification Participating organizations may receive a verification page on Marshall.net indicating that they are engaged with the framework. Verification pages may display: Organization name Participation status Framework version Reporting status Verification indicates participation in the framework rather than certification of compliance. --- # Founder Phase The Marshall AI Governance Standard is currently in the Founder Phase. During this phase the framework remains intentionally simple and evolves through practical use. Manual processes are acceptable while the framework matures. --- # Future Development Future versions of the framework may include: - expanded governance guidance - practitioner training - implementation tools - supporting documentation All changes will remain anchored to the Marshall Principle. --- Marshall AI Governance Standard Version 0.2 Founder Phase Published March 2026