CCAR-P Practice Question: Stakeholder Communication and Lifecycle Management
Your team operates a Claude-based internal knowledge assistant. A new compliance officer asks how the platform will handle model deprecations and capability changes over the next 24 months. Which TWO practices should be established now to give stakeholders durable lifecycle assurance? (Choose two.)
⚠ Common exam trap
The trap here is mistaking maximum freshness or a vendor permanence promise for lifecycle assurance, when assurance actually comes from controlled versioning plus documented ownership.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Pin production traffic to a specific model version and maintain a documented, tested upgrade path with a rollback option before any version change.
Durable lifecycle assurance rests on two complementary controls: version pinning with a rehearsed upgrade and rollback path, and a documented register that ties every model to its business function, owner, and review cadence. Pinning supplies predictability; the register supplies visibility and accountability. Promises of eternal model availability, constant upgrades to the newest version, and indefinite prompt freezes all fail because they either cannot be honoured or actively prevent the controlled change that deprecation management requires.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Ask the model provider to guarantee in writing that no model used by the platform will ever be retired.
Why it's wrong here
No provider can credibly guarantee perpetual availability of every model version; deprecation is a normal part of platform lifecycle management. Requesting such a guarantee wastes the compliance conversation and signals a misunderstanding of how hosted models evolve. The realistic assurance is a managed transition plan, not an impossible permanence commitment.
- ✓
Pin production traffic to a specific model version and maintain a documented, tested upgrade path with a rollback option before any version change.
Why this is correct
Pinning gives deterministic behaviour that compliance can reason about, while a tested upgrade path with rollback converts a disruptive event into a controlled change. Together they provide the predictability the officer is asking about, because the platform can demonstrate that a deprecation will not silently alter outputs in a regulated workflow.
- ✓
Publish a lifecycle register that maps each Claude model in use to its intended business function, known deprecation signals, owner, and scheduled review date.
Why this is correct
A lifecycle register makes the model inventory visible and accountable, which is exactly what a compliance officer needs to audit. By tying each model to a function, owner, and review cadence, it turns deprecation from an unmanaged surprise into a tracked obligation, and it creates the evidence trail that governance reviews require.
- ✗
Freeze all prompt and evaluation changes indefinitely so that the only variable in the system is the underlying model.
Why it's wrong here
While isolating variables has some testing value, an indefinite freeze prevents the platform from improving quality, fixing regressions, or adapting to model changes. It also does not address deprecation at all, since the provider will still retire versions. The result is a brittle system that cannot be maintained, which undermines rather than strengthens lifecycle assurance.
- ✗
Always route production requests to the newest available Claude model so the platform is never running outdated versions.
Why it's wrong here
Chasing the newest model maximizes change frequency and destroys output reproducibility, which is the opposite of lifecycle assurance. Compliance needs stable, auditable behaviour, not constant drift. This practice would make regression testing impossible because the baseline shifts continuously, and it exposes regulated workflows to untested capability changes without any rollback plan.
About these practice questions
One of 262 original CCAR-P practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Anthropic exam blueprint
This CCAR-P practice question is part of Courseiva's free Anthropic certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the CCAR-P exam.