Test your AI safety knowledge
AI safety is the field concerned with making advanced AI systems reliable, controllable, and aligned with human intent — spanning everything from near-term issues like bias and misuse to longer-horizon questions about powerful autonomous systems. This 25-question quiz checks your grasp of the core concepts: the alignment problem, training techniques like RLHF and constitutional AI, interpretability, and the foundations of AI governance. Every question comes with an explanation, so it doubles as a quick way to learn the vocabulary and the arguments.
How the quiz works
Answer one question at a time. Pick an option and submit to lock it in; the quiz then tells you whether you were right and explains the concept in plain language. Your score accumulates as you go, and at the end you see a total, a percentage, and a list of the topics you missed so you know where to read more. Everything runs in your browser — nothing is stored or sent anywhere — and you can retake it as many times as you like.
Tips for getting the most from it
Treat the explanations as the real payoff: even questions you answer correctly often clarify a distinction worth knowing — for example, the difference between alignment and capability, or between RLHF and constitutional AI. Where a topic is genuinely debated, the explanation flags that rather than pretending there is one settled answer. If you score lower than you hoped, the missed-topics list points you straight at the concepts to revisit, and a second attempt is the fastest way to make them stick.