What You Need to Know
Escalation passes a Claude output to a qualified person to review before use. Triage is the decision about who reviews each output. Triggers are properties of the situation that raise the cost of an error. This objective covers when your own review is not enough, and whose review is.
A driving school owner has four Claude drafts: a website advert, a refund reply under the booking terms, appraisal notes on an instructor, and a car-cleaning reminder. All four read well.
The Triggers
Each trigger below calls for qualified human review.
- an external audience: customers, the public, press, funders
- legal, financial or contractual consequence
- decisions about people: hiring, appraisal, allocation
- safety or health implications
- irreversibility once sent
One trigger is enough to warrant a qualified second pair of eyes. Several together make escalation non-negotiable. Confidence and polish are not on the list, because every Claude draft already has them. Claude is not a single source of truth, and high-stakes output deserves careful scrutiny.
At the driving school, the advert has an external audience. The refund reply has a contractual consequence. The appraisal notes feed a decision about a person. The cleaning reminder carries none, so the owner reviews it herself.
Outputs with no trigger get your own careful review, using the habits from objectives 2.1 to 2.3. These are internal working material, first drafts that will iterate, and routine low-consequence messages. That review is real, not a waiver: triage decides who reviews, never whether anyone does.
Qualified, Not Senior
Anthropic's Usage Policy sets the standard. It names High-risk Areas, such as legal, medical, financial and employment matters, where Claude's output can become advice to a person or a decision about them. The policy calls that output a High-risk AI Recommendation and requires two things for it: the person must be told AI was used, and a qualified person must review it. The review rule reads: "A qualified person must meaningfully review a High-risk AI Recommendation, with the authority to change or modify it, before it is delivered as advice". The same rule applies before it is used to make a decision. Advising someone on their legal rights, or on what to do in a specific legal matter, is one of these areas.
A qualified person is "someone with the training or experience to evaluate the output in that field". The words that pick the reviewer are "in that field". Contract language goes to someone with legal competence, tax figures to the accountant, clinical claims to the clinician, and hiring criteria to HR judgement. Outside the High-risk Areas the policy sets no review rule, but competence is still the test that picks the reviewer.
Seniority and availability are neither necessary nor sufficient. An unqualified reader, however senior, cannot catch the errors that make the content dangerous. A busy director skim-reading a lease clause gives comfort, not review. The refund reply goes to a solicitor, not to the owner.
Both Failure Directions
Under-escalation is the intuitive failure. It sends externally-bound content with professional-consequence claims because the draft reads authoritatively. Or it waves an output through as "only internal" when it feeds an external or consequential decision. Claude's output reads authoritatively by default, but the triggers belong to the situation and survive the polish. Where the output goes next decides more than where it was drafted. The appraisal notes stay in the office but feed a decision about an instructor.
Over-escalation feels safe because caution does. It is not. A committee for a rota note, or every routine draft routed through a professional reviewer, is an over-engineered detour. It buries reviewers, stalls routine work and teaches the organisation to ignore the escalation channel that high-stakes outputs need. A solicitor reading the cleaning reminder reaches the refund reply later. The skill is triage, not maximal caution.
Two Different Reasons to Stop
This objective escalates for quality and stakes: what happens if the output is wrong, and who is competent to catch it. Domain 6 covers other reasons to stop: the data is personal or regulated, or the use sits outside organisational AI policy. Both can arise in one situation, such as a refund reply quoting a pupil's personal data, but the trigger differs. The programme's technical escalation is a third kind. It hands a task to a Claude Developer or Architect because it needs automation or integration. That is the task outgrowing the chat, not an output needing a qualified reviewer.
In the exam
The exam tests this as triage: a decision on what the output carries, not how polished it looks. It penalises a wrong decision in either direction. In stems, look for the destination and for what qualifies a named reviewer. Candidates rarely expect over-escalation to be tested, and "escalate everything" is a real wrong answer. The exam keeps Domain 6 triggers distinct.
Key Concept
Triage asks two questions in order. What does this output carry? If anything, who is qualified in that subject to catch the errors that matter? Options answering with rank, availability or blanket process are the distractors.
Exam Traps
Choosing the reviewer by seniority or availability rather than subject competence
Where the Usage Policy requires review, for advice or a decision in one of its High-risk Areas, the reviewer must be a qualified person: someone with the training or experience to evaluate the output in that field. Competence is the right test outside those areas too. A senior manager's sign-off on legal wording or clinical claims is comfort, not review, so the keyed answer names competence: the person who can actually catch the dangerous error.
Waving output through because it is only internal
The trigger is what the output carries, not where it first lands. An internal analysis that feeds an external bid or a decision about people carries the consequence with it, so the keyed answer escalates on the destination and the stakes, not the label.
Escalating everything to be safe
Over-escalation is a tested wrong answer, not a cautious right one. Routing routine internal drafts through committees or professionals buries reviewers and erodes the channel high-stakes outputs need; the keyed answer lets careful self-review suffice for genuinely low-stakes material.
Practice Scenario
An opticians' practice manager has three Claude drafts: advertising copy for a new lens range claiming specific visual outcomes, a reworded internal rota-swap policy, and a reply to a patient complaint that touches on what may have gone wrong clinically. The manager has no clinical training and does not know what eye-care advertising may legally promise. Each draft must be checked by someone who knows its subject, with no needless delay. What is the right triage?
Build Exercise
Triage Your Last Five Outputs
What you'll learn
- How to spot the triggers that demand qualified review
- How to name a reviewer by subject competence rather than rank
- How to check a triage for both failure directions
- How to turn one triage session into a standing escalation list
- List the last five things Claude drafted for you, or generate five typical ones from your role
Why: Triage only becomes a habit when practised on the actual mix of stakes your role produces.
You should see: Five one-line descriptions of real outputs, from routine to consequential.
- For each output, mark the triggers present: external audience, legal or financial consequence, decisions about people, safety or health implications, irreversibility once sent
Why: The triggers are properties of the situation, and listing them stops the polish of the prose deciding the triage.
You should see: Each output with its triggers marked, or marked clean.
- For each triggered item, name the qualified reviewer in your organisation: a person or role competent in that subject, not merely senior
Why: Naming a real person tests whether the escalation channel actually exists before you need it.
You should see: A named reviewer per triggered item, with the competence that qualifies them.
- For each untriggered item, write down what your own careful review covers, using the objective 2.1 two-pass habit as the checklist
Why: Self-review is the correct outcome for low-stakes work, and writing it down keeps it a real review rather than a waiver.
You should see: A short self-review note per untriggered item.
- Check the finished triage for both failure directions, then keep the result as your standing escalation list
Why: Under- and over-escalation are both tested, and both show up clearly in a written triage.
You should see: Nothing external-facing waved through, nothing routine queued for committee, and a reusable list.
Sources
- Usage Policy — Anthropic
- Claude is providing incorrect or misleading responses. What's going on? — Anthropic
- Claude Certified Associate – Foundations Exam Guide — Anthropic