Skip to content
AI Investigators · Bias in AI

Before we start: quick check

3 quick questions. No account and no name needed. Your answers are counted anonymously to help ARK see what workshops teach.

Question 1 of 3: An AI tool learns to spot spam by studying lots of emails that people already labeled "spam" or "not spam." What are those labeled emails called?
Question 2 of 3: A company trains an AI tool on its past hiring decisions. In the past, it mostly hired people from one group. What is the tool likely to do?
Question 3 of 3: Which step does the most to catch bias before an AI tool is used on real people?

Only used to count answers by grade. Skip it if you like.