{"id":794,"date":"2026-06-16T16:49:29","date_gmt":"2026-06-16T16:49:29","guid":{"rendered":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/chapter\/how-ai-learns\/"},"modified":"2026-06-16T16:49:29","modified_gmt":"2026-06-16T16:49:29","slug":"how-ai-learns","status":"publish","type":"chapter","link":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/chapter\/how-ai-learns\/","title":{"raw":"How AI Learns","rendered":"How AI Learns"},"content":{"raw":"\nShow <strong>slide seven<\/strong>, or project <a href=\"https:\/\/textbook.mediasmarts.ca\/blockswrdsb\/chapter\/is-this-a-cat\/\"><em>Is This a Cat?<\/em><\/a>\n<div class=\"textbox shaded\">\n\nPick a student volunteer to be the \u201cAI robot.\u201d Tell them the secret rule: <strong>everything that is orange is a cat<\/strong>, and <strong>anything that is not orange is not a cat<\/strong>.\n\n<\/div>\nTell the other students that the AI has been told the first three images are all cats.\n\nAsk them to guess:\n\n&nbsp;\n<ul>\n \t<li>Will it say the black cat is a cat?<\/li>\n \t<li>Will it say the fox is a cat?<\/li>\n<\/ul>\n<div class=\"textbox shaded\">\n\nHave the \u201cAI\u201d volunteer give the answer: the black cat is not a cat, and the fox is a cat.\n\n<\/div>\n&nbsp;\n\nShow <strong>slide eight <\/strong>if you are using the slides.\n\nAsk:\n<ul>\n \t<li>Is that the right answer? <em>(No. A black cat is a cat, and a fox is not a cat.)<\/em><\/li>\n \t<li>Would a person give that answer? (<em>Probably not! After seeing three cats, you would almost certainly be able to tell the difference between a cat and a fox.)<\/em><\/li>\n<\/ul>\nShow <strong>slide nine <\/strong>if you are using the slides.\n\nAsk:\n\n&nbsp;\n<ul>\n \t<li>Why do you think the \u201crobot\u201d said that?<\/li>\n \t<li>What rule was it following?<\/li>\n<\/ul>\n<div class=\"textbox shaded\">\n\nLet students discuss for a few minutes. Don\u2019t try to reach any sort of consensus or definition yet.\n\n<\/div>\nShow <strong>slide ten <\/strong>if you are using the slides.\n<div class=\"textbox shaded\">\n\nTell students that the reason AIs seem like they \u201cthink\u201d and \u201cdecide\u201d is because they look for patterns in what their makers show them. They test those patterns and the ones that work become rules.\n\n<\/div>\n&nbsp;\n\n&nbsp;\n\nLet the \u201cAI robot\u201d student explain what the rule was:\n<div class=\"textbox shaded\">\n\nEverything that is orange is a cat, and anything that is not orange is not a cat.\n\n<\/div>\n&nbsp;\n\nShow <strong>slide eleven <\/strong>if you are using the slides.\n\nAsk students:\n<ul>\n \t<li>How did it make this rule? <em>(All of the cats it was shown were orange.)<\/em><\/li>\n \t<li>How would you change this rule so that it would recognize that black cats were cats, and that foxes were not cats?<\/li>\n<\/ul>\nLet students discuss this for a few minutes. Make sure to point out that you can <em>correct <\/em>an AI by telling it what answers are right or wrong.\n<div class=\"textbox shaded\">\n\nExplain that you can do this in two ways: by having its makers test and correct it <em>before <\/em>people start using it, or by having the AI <em>correct itself <\/em>each time it\u2019s wrong. (Most AIs do both!)\n\n<\/div>\n&nbsp;\n\nShow <strong>slide twelve <\/strong>if you are using the slides, or take out the bouncing balls.\n\n&nbsp;\n<div class=\"textbox shaded\">\n\nTell the student volunteer the secret rule: <strong>everything that is round bounces.<\/strong>\n\n<\/div>\n&nbsp;\n\nAsk students: If the first three items all bounce, will the AI say that the last one will bounce?\n\n&nbsp;\n<div class=\"textbox shaded\">\n\nHave the \u201cAI\u201d volunteer give the answer: the last ball will bounce.\n\n<\/div>\n&nbsp;\n\nShow <strong>slide thirteen <\/strong>if you are using the slides, or take out one of the round fruit.\n\n&nbsp;\n\nAsk students: Will the AI say that the fruit will bounce?\n\n&nbsp;\n<div class=\"textbox shaded\">\n\nHave the \u201cAI\u201d volunteer give the answer: the fruit will bounce.\n\n<\/div>\nShow <strong>slide fourteen <\/strong>if you are using the slides.\n\nAsk students:\n<ul>\n \t<li>What pattern did the AI spot?<\/li>\n \t<li>What rule did it make?<\/li>\n \t<li>Why did it give the right answer once?<\/li>\n<\/ul>\nWhy did it give the wrong answer once?\n<div class=\"textbox shaded\">\n\nLet the \u201cAI robot\u201d student explain what the rule was:\n\nEverything that is round bounces.\n\n<\/div>\nExplain that because that we don\u2019t always know when AI makes a mistake, because (as in this case) it may give the right answer for the reason.\n\nShow <strong>slide fifteen <\/strong>if you are using the slides.\n\nAsk students: If you showed an AI a basketball, a volleyball, and an orange, would it guess that a soccer ball would bounce? Would it guess that an apple would bounce?\n\nLet students discuss this for a few minutes. Make sure to point out that you can make an AI give better answers by <em>showing it more examples <\/em>so it doesn\u2019t make rules that are too narrow (for instance, showing it cats that are not all orange, or that not all round things bounce.)\n<div class=\"textbox shaded\">\n\nExplain that you can do this in two ways: by showing the AI more examples <em>before <\/em>people start using it, or by having the AI use every right or wrong guess as a new example. (Most AIs do both!)\n\n<\/div>\n&nbsp;\n","rendered":"<p>Show <strong>slide seven<\/strong>, or project <a href=\"https:\/\/textbook.mediasmarts.ca\/blockswrdsb\/chapter\/is-this-a-cat\/\"><em>Is This a Cat?<\/em><\/a><\/p>\n<div class=\"textbox shaded\">\n<p>Pick a student volunteer to be the \u201cAI robot.\u201d Tell them the secret rule: <strong>everything that is orange is a cat<\/strong>, and <strong>anything that is not orange is not a cat<\/strong>.<\/p>\n<\/div>\n<p>Tell the other students that the AI has been told the first three images are all cats.<\/p>\n<p>Ask them to guess:<\/p>\n<p>&nbsp;<\/p>\n<ul>\n<li>Will it say the black cat is a cat?<\/li>\n<li>Will it say the fox is a cat?<\/li>\n<\/ul>\n<div class=\"textbox shaded\">\n<p>Have the \u201cAI\u201d volunteer give the answer: the black cat is not a cat, and the fox is a cat.<\/p>\n<\/div>\n<p>&nbsp;<\/p>\n<p>Show <strong>slide eight <\/strong>if you are using the slides.<\/p>\n<p>Ask:<\/p>\n<ul>\n<li>Is that the right answer? <em>(No. A black cat is a cat, and a fox is not a cat.)<\/em><\/li>\n<li>Would a person give that answer? (<em>Probably not! After seeing three cats, you would almost certainly be able to tell the difference between a cat and a fox.)<\/em><\/li>\n<\/ul>\n<p>Show <strong>slide nine <\/strong>if you are using the slides.<\/p>\n<p>Ask:<\/p>\n<p>&nbsp;<\/p>\n<ul>\n<li>Why do you think the \u201crobot\u201d said that?<\/li>\n<li>What rule was it following?<\/li>\n<\/ul>\n<div class=\"textbox shaded\">\n<p>Let students discuss for a few minutes. Don\u2019t try to reach any sort of consensus or definition yet.<\/p>\n<\/div>\n<p>Show <strong>slide ten <\/strong>if you are using the slides.<\/p>\n<div class=\"textbox shaded\">\n<p>Tell students that the reason AIs seem like they \u201cthink\u201d and \u201cdecide\u201d is because they look for patterns in what their makers show them. They test those patterns and the ones that work become rules.<\/p>\n<\/div>\n<p>&nbsp;<\/p>\n<p>&nbsp;<\/p>\n<p>Let the \u201cAI robot\u201d student explain what the rule was:<\/p>\n<div class=\"textbox shaded\">\n<p>Everything that is orange is a cat, and anything that is not orange is not a cat.<\/p>\n<\/div>\n<p>&nbsp;<\/p>\n<p>Show <strong>slide eleven <\/strong>if you are using the slides.<\/p>\n<p>Ask students:<\/p>\n<ul>\n<li>How did it make this rule? <em>(All of the cats it was shown were orange.)<\/em><\/li>\n<li>How would you change this rule so that it would recognize that black cats were cats, and that foxes were not cats?<\/li>\n<\/ul>\n<p>Let students discuss this for a few minutes. Make sure to point out that you can <em>correct <\/em>an AI by telling it what answers are right or wrong.<\/p>\n<div class=\"textbox shaded\">\n<p>Explain that you can do this in two ways: by having its makers test and correct it <em>before <\/em>people start using it, or by having the AI <em>correct itself <\/em>each time it\u2019s wrong. (Most AIs do both!)<\/p>\n<\/div>\n<p>&nbsp;<\/p>\n<p>Show <strong>slide twelve <\/strong>if you are using the slides, or take out the bouncing balls.<\/p>\n<p>&nbsp;<\/p>\n<div class=\"textbox shaded\">\n<p>Tell the student volunteer the secret rule: <strong>everything that is round bounces.<\/strong><\/p>\n<\/div>\n<p>&nbsp;<\/p>\n<p>Ask students: If the first three items all bounce, will the AI say that the last one will bounce?<\/p>\n<p>&nbsp;<\/p>\n<div class=\"textbox shaded\">\n<p>Have the \u201cAI\u201d volunteer give the answer: the last ball will bounce.<\/p>\n<\/div>\n<p>&nbsp;<\/p>\n<p>Show <strong>slide thirteen <\/strong>if you are using the slides, or take out one of the round fruit.<\/p>\n<p>&nbsp;<\/p>\n<p>Ask students: Will the AI say that the fruit will bounce?<\/p>\n<p>&nbsp;<\/p>\n<div class=\"textbox shaded\">\n<p>Have the \u201cAI\u201d volunteer give the answer: the fruit will bounce.<\/p>\n<\/div>\n<p>Show <strong>slide fourteen <\/strong>if you are using the slides.<\/p>\n<p>Ask students:<\/p>\n<ul>\n<li>What pattern did the AI spot?<\/li>\n<li>What rule did it make?<\/li>\n<li>Why did it give the right answer once?<\/li>\n<\/ul>\n<p>Why did it give the wrong answer once?<\/p>\n<div class=\"textbox shaded\">\n<p>Let the \u201cAI robot\u201d student explain what the rule was:<\/p>\n<p>Everything that is round bounces.<\/p>\n<\/div>\n<p>Explain that because that we don\u2019t always know when AI makes a mistake, because (as in this case) it may give the right answer for the reason.<\/p>\n<p>Show <strong>slide fifteen <\/strong>if you are using the slides.<\/p>\n<p>Ask students: If you showed an AI a basketball, a volleyball, and an orange, would it guess that a soccer ball would bounce? Would it guess that an apple would bounce?<\/p>\n<p>Let students discuss this for a few minutes. Make sure to point out that you can make an AI give better answers by <em>showing it more examples <\/em>so it doesn\u2019t make rules that are too narrow (for instance, showing it cats that are not all orange, or that not all round things bounce.)<\/p>\n<div class=\"textbox shaded\">\n<p>Explain that you can do this in two ways: by showing the AI more examples <em>before <\/em>people start using it, or by having the AI use every right or wrong guess as a new example. (Most AIs do both!)<\/p>\n<\/div>\n<p>&nbsp;<\/p>\n","protected":false},"author":2,"menu_order":3,"template":"","meta":{"pb_show_title":"on","pb_short_title":"","pb_subtitle":"","pb_authors":[],"pb_section_license":""},"chapter-type":[],"contributor":[],"license":[],"class_list":["post-794","chapter","type-chapter","status-publish","hentry"],"part":791,"_links":{"self":[{"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/pressbooks\/v2\/chapters\/794","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/pressbooks\/v2\/chapters"}],"about":[{"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/wp\/v2\/types\/chapter"}],"author":[{"embeddable":true,"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/wp\/v2\/users\/2"}],"version-history":[{"count":0,"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/pressbooks\/v2\/chapters\/794\/revisions"}],"part":[{"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/pressbooks\/v2\/parts\/791"}],"metadata":[{"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/pressbooks\/v2\/chapters\/794\/metadata\/"}],"wp:attachment":[{"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/wp\/v2\/media?parent=794"}],"wp:term":[{"taxonomy":"chapter-type","embeddable":true,"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/pressbooks\/v2\/chapter-type?post=794"},{"taxonomy":"contributor","embeddable":true,"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/wp\/v2\/contributor?post=794"},{"taxonomy":"license","embeddable":true,"href":"https:\/\/textbook.mediasmarts.ca\/blocks-teachertext\/wp-json\/wp\/v2\/license?post=794"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}