Containment
The set of controls that bound how much harm an agent can do while running, regardless of what it decides. Containment includes scoped and revocable credentials, spend caps, rate caps, allowlists, sandboxing, approval gates on irreversible actions, and staged rollout. Its defining test is a specific, survivable answer to "what is the worst this agent can do in one hour if every judgment it makes is wrong."
Defined in 3 GAGE programs, which carry 7 distinct definitions of it. The wording above is taught in Certified AI Governance Professional (CAIGP).
How each discipline defines it
The same term does different work depending on who is using it. These are the definitions as each program teaches them, unedited.
The set of controls that bound how much harm an agent can do while running, regardless of what it decides. Containment includes scoped and revocable credentials, spend caps, rate caps, allowlists, sandboxing, approval gates on irreversible actions, and staged rollout. Its defining test is a specific, survivable answer to "what is the worst this agent can do in one hour if every judgment it makes is wrong."
The first remediation layer, stopping the scheme now while preserving the evidence. Decided with the investigation lead and counsel, because speed here can destroy the recovery case.
The immediate technical and operational response that stops an incident from getting worse while the rest of the response proceeds.
The action that stops the harm, typically executing the pre-tested rollback or off switch from the launch plan. It stops new harm but cannot recall harm already delivered, and it is the pivot of the response, not its end.
The graduated, live response to a runaway while the agent keeps running, a ladder of moves (detect, throttle, cap the loop, narrow the scope, isolate, kill) between "let it run" and "pull the plug." Distinct from the kill switch, which is only the top rung.
Where it is taught
The exact lessons this term appears in. The first module of every program is free with a free account.
- The 2 a.m. incident: your system fails live, stakeholders are calling, run the response · Shipping AI and Surviving the Incident, Certified AI Governance Professional (CAIGP)
- You deploy an agent: scoping, containment, and the kill switch you test before launch · Agents Under Command, Certified AI Governance Professional (CAIGP)
- The runaway afternoon: your agent misbehaves in a loop, contain it live · Agents Under Command, Certified AI Governance Professional (CAIGP)
- Lab: Incident Response Drill · Enforcement and Incident Response, Certified China AI Regulatory Professional (CCARP)
- Incident management and crisis communication · Cybersecurity Governance and Operational Resilience, Certified Enterprise Governance, Risk and Compliance Professional (CEGP)
- Remediation, discipline, and recovery · Financial Risk, Fraud, and Financial Crime, Certified Enterprise Governance, Risk and Compliance Professional (CEGP)
- Remediation design · Findings, Issues, Root Cause, and Remediation, Certified Enterprise Governance, Risk and Compliance Professional (CEGP)
Where it shows up at work
A definition is where Containment starts. The lessons above teach it inside the competencies below, which are what job descriptions ask for and what the GAGE credential records. Every link here is drawn from the same careers map that builds a role's readiness check.
Classifies AI incidents by severity, runs containment, preserves evidence, manages notification, and closes the loop with lessons learned.
2 lessons on this page teach it as a primary outcome.
Builds minimization, purpose limitation, de-identification, consent and retention into an AI system before launch, with tests that prove it.
One lesson on this page teach it as a primary outcome.
Designs multi step agent systems: orchestration graphs, tool contracts, memory, interruptibility, and the evaluation of agent behavior before it is allowed to act.
One lesson on this page teach it as a primary outcome.
Roles that hire for it
Each role page shows the full competency set, what the seat decides, and a readiness check against your own evidence.
- Chief Privacy Officer, AI privacy editionneeds it at depth
- Chief Information Security Officer, AI security focusneeds it at depth
- AI Incident Response Leadneeds it at depth · 1 open posting this week
- Chief Risk Officer, AI risk editionneeds working knowledge
- Agentic AI Engineer, orchestration editionneeds it at depth
- AI Security Architectneeds working knowledge · 8 open postings this week
How job descriptions say it
Postings rarely use the textbook word. These are the phrases the careers map reads as asking for the competencies above.
Where the frameworks place it: EU AI Act Article 73; ISO/IEC 42001 Annex A incident controls; NIST AI RMF MANAGE 4; ISO/IEC 27701.
One interview question about it
An AI system has just caused harm to a customer. Walk me through the first 48 hours.
Asked of candidates for ai incident response and recovery. Answering it in your own words is exactly how a GAGE topic is passed, and the credential records the pass where an employer can check it.
If you come from another field
Some backgrounds already carry this competency, and the readiness check gives credit for it.
- Cybersecurity and IT carries it as work you can show: Incident playbooks and post-incident reviews are your artifacts.
- A cybersecurity or information assurance degree carries it as a claim to back up: Incident response is a course; AI incidents add model failure to the playbook.
Terms it appears with
Not an alphabetical neighbourhood: these are the terms taught in the same lessons, ranked by how often they appear together.