Here is the honest situation. Here is the honest situation. When an AI system starts behaving in a way no one intended, an engineer can roll it back in minutes, but the questions that decide whether your organization is exposed are governance questions. Who is authorized to stop it. On whose signature. What has to be true for that authority to fire. What record proves the decision was made properly and in time. Whether stopping it cleanly was even possible, or whether the only options were letting harm continue or taking down a service the business depends on. Doing this well means designing an emergency shutdown and rollback procedure a regulator would accept, one that names its triggers, its authority, its safe state and its record. It means placing kill authority in a named human with the real power to act and the seniority to act without fear. It means building a theory-of-harm that traces a plausible path from a model output to a party harmed, and allocating that exposure across the operator, the provider and third-party model suppliers. It means balancing operational continuity against safety by the severity of the harm, keeping a contemporaneous incident record, knowing your reporting triggers before the clock starts, structuring accountable ownership, and exercising the whole thing before it is real. Where compliance functions fall short is predictable: a shutdown that exists only as an engineer's kill switch, oversight that is a title rather than command, liability treated as vague dread, and mandate compliance held as a binder of policies rather than obligations mapped to operating controls.
This Kit removes the guesswork. It is AI governance and emergency response written as adopt-ready controls you personalize in a weekend, with the evidence a compliance, risk or regulatory reviewer examines.
What you get, the moment you buy
Grounded in AI governance practice and the recurring obligations of the emerging AI safety mandates. Editable Word and Excel files.
What one control looks like
This is the opening control, where accountable ownership is set. All 18 are built to this depth.
Why this is not another template pack
- The evidence is the point. A control you cannot evidence is a gap waiting to be found. This tells you what a compliance officer, risk manager or regulator examines and where teams fall short, for every control.
- The emergency specifics built in. Shutdown triggers and safe states, named kill authority and human command, the theory-of-harm and liability allocation, the continuity-versus-safety mapping, the contemporaneous incident record and reporting triggers are written into the controls, not left generic.
- Built on real practice, not one person's opinion, grounded in how AI safety mandates translate into what an organization actually has to be able to do when an autonomous system has to be stopped.
- It compounds. This work shares its shape with AI risk management, incident response and regulatory reporting, so it feeds your wider governance and compliance practice.
Who buys this
Compliance officers, risk managers and legal counsel overseeing AI system deployment and regulatory adherence, and the governance owners of the shutdown procedures, kill authority and incident records those AI systems depend on. Whether this is your first AI emergency procedure or a maturity uplift, you save weeks and walk in with your shutdown, kill-authority, liability, continuity, incident-record and evidence controls structured.
Common questions
Is it really editable? Yes. Word and Excel files you own and adapt. No portal, no subscription.
Does it cover the full emergency arc? Yes. Governance foundation and mandate mapping, emergency shutdown and rollback design, kill authority and human command, liability and theory-of-harm, continuity, incident record and regulatory reporting, and exercising and evidencing compliance each have their own controls with their own evidence.
Is this tied to one regulation or jurisdiction? No. The controls are principle-level and map to the recurring obligations across instruments such as the EU AI Act, the NIST AI Risk Management Framework and ISO/IEC 42001, so they apply whichever mandates reach your AI systems.
What if it is not for me? A 30-day money-back guarantee.
Instant digital download · 30-day money-back guarantee · The Art of Service Pty Ltd, GPO Box 2673, Brisbane QLD 4001 · support@theartofservice.com