GH-300 domain - 18% of the exam

Use GitHub Copilot responsibly

Use GitHub Copilot responsibly is 18% 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 sampleUse GitHub Copilot responsiblyeasy

A developer accepts a GitHub Copilot suggestion that calls a date-parsing helper, and the function name and signature look entirely convincing. Which risk of generative AI tools does this scenario most directly illustrate?

  • ABias, where the suggestion reflects skewed patterns in the training data and treats one group of inputs less fairly than another group.
  • BVendor lock-in, where relying on Copilot ties the developer to a single tool and makes switching to another assistant prohibitively costly later.
  • CLatency, where the suggestion arrives too slowly to be useful and interrupts the developer's flow while writing the date-parsing routine.
  • DHallucination, where the model produces a confident, plausible-looking suggestion that may reference an API or behaviour that does not actually exist. Correct
Identify hallucination as the risk where a generative AI tool produces confident output that may reference things that do not exist. Hallucination occurs when a language model generates fluent, confident output that is not grounded in fact, such as a plausible function or API that does not exist, which is why a convincing suggestion still has to be validated rather than trusted on appearance.

Why A is wrong: Bias concerns unfair treatment learned from skewed data, but a plausibly named helper that may not exist is about fabricated output rather than unfairness.

Why B is wrong: Lock-in is a commercial concern about switching tools, which has nothing to do with whether a specific suggested helper actually exists or behaves correctly.

Why C is wrong: Latency is about response speed, not correctness, so it does not describe a suggestion that looks right but may reference something that is not real.

Why D is correct: Hallucination is exactly a confident, plausible output that may invent a non-existent API, so a convincing helper that might not exist is the textbook example.

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.