Artificial Intelligence MCQ

Q: The 'context window' of a language model determines:

Correct answer: How much text it can consider at one time. The context window limits how much of the conversation and documents the model can attend to at once.

Q: Diffusion models generate images by:

Correct answer: Iteratively removing noise guided by a text prompt. Diffusion starts from random noise and denoises step by step toward an image matching the prompt.

Q: Which request is MOST likely to produce a hallucination?

Correct answer: List the exact page numbers of quotes in a book you have not been given. Specific verifiable details the model has no access to — citations, page numbers — are prime hallucination territory.

Q: RLHF (reinforcement learning from human feedback) is used to:

Correct answer: Align model behaviour with human preferences. RLHF fine-tunes models using human preference ratings, making outputs more helpful and safe.

Q: What should you do before pasting a work document into a public AI chatbot?

Correct answer: Check confidentiality policies and remove sensitive data. Public tools may retain inputs; confidential data should never be shared without approval.

AI concepts, machine learning, generative AI and prompt engineering.

Q21. The 'context window' of a language model determines: Medium

Permalink →

Q22. Diffusion models generate images by: Medium

Permalink →

Q23. Which request is MOST likely to produce a hallucination? Medium

Permalink →

Q24. RLHF (reinforcement learning from human feedback) is used to: Medium

Permalink →

Q25. What should you do before pasting a work document into a public AI chatbot? Medium

Permalink →

Q26. A well-structured prompt typically includes: Medium

Permalink →

Q27. Why do LLMs sometimes state falsehoods confidently? Medium

Permalink →

Q28. Which is an ethically sound academic use of generative AI? Medium

Permalink →

Q29. AI voice cloning creates which new security risk? Medium

Permalink →

Q30. The most reliable habit for handling AI-generated factual claims is: Medium

Permalink →

🎯 Want a real test?

Take a timed quiz on this topic and see your rank on the leaderboard.

Browse Quizzes