GH-300 domain - 13% of the exam

Understand GitHub Copilot data and architecture

Understand GitHub Copilot data and architecture is 13% of the GitHub Copilot (GH-300) exam. These are the objectives it covers, each with practice questions and worked explanations.

Objectives in this domain

Sample question from this domain

Free sampleUnderstand GitHub Copilot data and architecturemedium

A compliance lead at a company on Copilot Business is briefing legal on how developer data is handled by GitHub Copilot during everyday coding. They want only statements that are accurate for the Copilot Business and Copilot Enterprise arrangement. Which two statements should they include? Select TWO.

  • ABecause the business tiers are used, the proxy guarantees every returned suggestion is correct, secure, and free of any licensing concern before it reaches the editor.
  • BOn the business tiers, developer prompts are always stored indefinitely on the local workstation so the same suggestion can be replayed offline at any later time.
  • CPrompts and suggestions generated by their developers are not retained to train the underlying foundation models under the Copilot Business and Enterprise arrangement. Correct
  • DChoosing Copilot Business automatically removes the developer's duty to review generated code, since the organisation contract transfers that responsibility to GitHub.
  • EThe Copilot proxy can filter and post-process a candidate suggestion before it is shown, so toxic content and certain public-code matches may be screened out. Correct
For Copilot Business and Enterprise, prompts and suggestions are not used for model training and the proxy filters candidates, yet no guarantees of correctness arise. Under the Copilot Business and Copilot Enterprise arrangement, developer prompts and suggestions are not used to train the foundation models, and the proxy filters and post-processes candidates to screen toxic content and public-code matches; however, post-processing never certifies correctness, security, or licence, retention is not indefinite local storage, and the developer keeps accountability for reviewing output.

Why A is wrong: Tempting because the proxy does post-process, but it never guarantees correctness, security, or licence cleanliness, and the developer remains accountable for the output.

Why B is wrong: Tempting given retention questions arise, but this invents indefinite local storage and offline replay, which is not how the data flow or retention works.

Why C is correct: It is accurate that prompts and suggestions for Copilot Business and Copilot Enterprise are not used to train the underlying foundation models, which is the assurance legal needs.

Why D is wrong: Tempting because the tier adds controls, but accountability for reviewing output stays with the developer on every plan and is not transferred by the contract.

Why E is correct: The proxy does filter and post-process candidates, screening for toxic content and, when enabled, suggestions matching public code, which is an accurate part of the data flow.

Other domains in this exam

See also the GH-300 cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by GitHub. Original, blueprint-aligned practice material only.