What types of prompts or code snippets might be flagged by the GitHub Copilot toxicity filter? (Each correct answer presents part of the solution. Choose two.)

The correct answer is A. Hate speech or discriminatory language (e.g., racial slurs, offensive stereotypes) B. Sexually suggestive or explicit content. GitHub Copilot includes a toxicity filter to prevent harmful or offensive content in generated suggestions. This filter targets hate speech or discriminatory language (A) and sexually suggestive or explicit content (B), both of which violate responsible AI usage policies…

Responsible AI

Question

Options

AHate speech or discriminatory language (e.g., racial slurs, offensive stereotypes)
BSexually suggestive or explicit content
CCode that contains logical errors or produces unexpected results
DCode comments containing strong opinions or criticisms

How the community answered

(37 responses)

A
89% (33)
C
3% (1)
D
8% (3)

Explanation

GitHub Copilot includes a toxicity filter to prevent harmful or offensive content in generated suggestions. This filter targets hate speech or discriminatory language (A) and sexually suggestive or explicit content (B), both of which violate responsible AI usage policies. Option C (logical errors) is a code quality issue, not toxicity - that's handled by developer review, not a toxicity filter. Option D (strong opinions in comments) does not meet the threshold for toxicity filtering.

Topics

#GitHub Copilot safety#Toxicity filter#Content moderation#Responsible AI

Community Discussion

No community discussion yet for this question.

Full GH-300 Practice