What is a Human of Record?
A Human of Record is the named person who answers for an AI agent: what it can reach, what it does and the decisions made about it. The term is the one Guard0 uses. It is assigned per agent to an individual, not to a team alias, and it moves to a new person when the owner leaves.
Also called agent owner, AI agent ownership, accountable owner. Last reviewed .
Why it matters
Law and frameworks already point at a person. Article 26 of the EU AI Act requires deployers of high-risk systems to "assign human oversight to natural persons who have the necessary competence, training and authority". NIST's GOVERN 2.1 asks that roles and responsibilities for AI risk be documented and clear.
An organization cannot assign the responsibility to the software. In Moffatt v. Air Canada, 2024 BCCRT 149, decided in February 2024, Air Canada argued that its website chatbot was a separate legal entity responsible for its own actions. The tribunal rejected the argument and held the airline liable for what the chatbot told a customer.
Related terms
- AI governance
- AI governance is the programme that requires every agent to have an owner.
- AI agent inventory
- An AI agent inventory is where the Human of Record is written down.
- Decision record
- A decision record names the Human of Record beside each decision.
- Shadow AI
- Shadow AI is, by definition, AI with no Human of Record.
How Guard0 handles it
Every agent on the Guard0 register has a Human of Record. Makers are resolved through single sign-on, findings and approvals go to that person, and when someone leaves their agents are listed in offboarding and move to a new owner.
Sources
- Regulation (EU) 2024/1689 (the EU AI Act). EUR-Lex.
- AI RMF Playbook: Govern. NIST.
- Moffatt v. Air Canada, 2024 BCCRT 149. British Columbia Civil Resolution Tribunal, via CanLII.