結論：NEW TREASURE（ニュートレジャー）Stage 2 Lesson 11のReadテーマである「AIと人間の協働」では、AIが提案した内容と確認できた事実を区別し、学校のルール・個人情報・出典を守って英語で説明することが重要です。文部科学省の学校向け資料でも、人間中心の利活用と児童生徒の発達段階・リスクへの配慮が示されています。慶英会は架空の授業を題材にした英文2題・独自32問・全問の根拠解説を無料公開します。
慶英会 塾長
慶英グループ CEO
一般選抜にて慶應義塾大学合格（英語入試正答率97％）
成績優秀者（慶應義塾維持会奨学生）として
慶應義塾大学を卒業（2023年）
一般財団法人 慶應連合三田会
慶應義塾横浜三田会 会員
高校在学中に英検1級取得
公益財団法人 日本英語検定協会「成績優秀者」に選出
中高主要科目で「オール5」
（最高評価・高校偏差値70）
国内最高峰の英語スピーチコンテスト
「高円宮杯全日本中学校英語弁論大会」
横浜市大会入賞・神奈川県大会進出
米国高校留学中、唯一の日本人として
全米最大級の高校生リーダーシップ団体
「FBLA」主催の英語スピーチコンテストに出場
ニューヨーク州大会準優勝・全米大会進出
米国の在籍高校代表として
「シカゴ国際模擬国連会議」に出場
慶應義塾大学在学中より英語指導を行い
「英語個別指導オンライン塾｜慶英会」を設立
小学生から、現役英語講師・英検面接官を含む
社会人まで数百名以上を指導
（最前列左から2人目：塾長）
Ｚ会の公式2026年カタログではStage 2 Lesson 11のReadにAIと人間の協働を扱う説明文が位置づけられています。本ページは共通するテーマを用いた慶英会独自教材で、教科書本文やAI製品の実際の会話を転載していません。学校での生成AI利用が認められているとは限らないため、担当の先生の指示を優先してください。
AIと人間が働くときの4つの確認事項
- 出典：日時・数値・歴史的事実を確認し、公式情報が必要なら原典を調べる。
- プライバシー：生徒の成績、氏名、顔写真や家庭事情をむやみに入力・公開しない。
- 学校の決まり：利用できるツールや課題の条件は学校・教員の指示に従う。
- 学習者の責任：提出する内容の意味を自分で説明し、検証可能な主張と仮説を分ける。
文部科学省の学校向け生成AI資料
文部科学省の学校現場における生成AI利用案内は、人間中心の利用を基本にし、学習活動における児童生徒の利用について発達段階・情報活用能力・リスクへの対策を踏まえて検討すべきと説明しています。学校が利用を認めていない課題にツールを利用することは避け、適切なルールを確認しましょう。
Lesson 11の英文法にもつなげる
The teacher asked us to check the source. はask＋人＋to不定詞、Could you tell me where the information came from? は間接疑問、The claim was too unreliable for us to use. はtoo … to不定詞です。文法の正しさと情報の正確性を両方確認します。
問題編｜架空の授業を読む2題・全32問
Read 1｜Who Checks the Answer?（379語）
At a fictional school, students prepared an English guide about places near their town. A teacher asked them to use reliable information and check names, dates, and opening hours. The teacher also reminded students that private information about classmates should not be placed in an online tool or published without permission. Each group was responsible for explaining where its information came from.
One group was allowed to use a generative AI tool only to suggest general research questions under the teacher’s supervision. The students asked it for possible topics, then searched for verified information through official sources. The tool produced a fluent paragraph about the local museum. However, the paragraph claimed that the museum was open on Mondays, while the official schedule used in the fictional class said it was closed. The students corrected the error and wrote down the source.
Another paragraph from the tool described the history of an old bridge. It sounded reasonable, but the group could not find reliable sources for two statements. Instead of copying them, the students checked an approved history book and wrote their own explanation. A student named Emi said that natural-sounding English and accurate facts were not the same thing.
Before the presentation, students reviewed their own words. They decided to say clearly which questions the tool had helped them think of and which facts they had verified themselves. They also checked that no private details about other students were included. The teacher asked them to explain why the museum schedule had been corrected.
At the final presentation, the group showed the official information alongside its original questions. Students could explain the corrections because they understood the difference between a suggestion and a verified fact. The teacher praised their careful work but did not claim that using AI always improves learning. The lesson was about responsible choices, not about proving one tool is better than every other way of studying.
Afterwards, Emi wrote a reflection. She thought digital tools could be useful when their use was permitted and carefully reviewed. Yet she understood that human responsibility remained important. If a claim was not supported by evidence, she should not repeat it as certain. She wanted to remember a simple question before publishing anything: ‘How do I know this is true?’
- Q1. What did the students prepare?
- Q2. Who set the rules about reliable sources and privacy?
- Q3. What limited use of an AI tool was allowed?
- Q4. What incorrect claim appeared in the draft?
- Q5. How did the students correct the museum information?
- Q6. Why did they avoid copying the bridge paragraph?
- Q7. What private details did they remove or avoid?
- Q8. Did the group submit the AI-generated paragraph unchanged?
- Q9. What did the tool actually help them do?
- Q10. Who was responsible for the final presentation?
- Q11. Explain the contrast introduced by ‘However’ in paragraph 2.
- Q12. Retell the main lesson in two English sentences.
Read 2｜A Presentation that Needed Revision（375語）
Another fictional English class prepared short presentations about food loss. The students were asked to make their opinions understandable and support factual claims with reliable sources. Their teacher said that school rules determined which tools could be used. Students were allowed to use dictionaries and approved study materials, but any use of generative AI required the teacher’s specific permission.
Team A began without an AI draft. Its members wrote three questions: What is food loss? Which figures are official estimates? What information would help us understand our own school? The team used a government page to check the definition of food loss and wrote a short explanation in English. When it used invented numbers for a practice table, it labeled them clearly as fictional data.
Team B first received a computer-generated paragraph. One sentence said that a new school plan would cut food loss in every school by exactly half. The group could not find evidence for that claim. One student wanted to use the sentence because it sounded persuasive, but another asked whether anyone had measured the result. The answer was no.
The students discussed how to revise the paragraph. They changed the unsupported promise into a more careful suggestion: checking food before buying more might help in some situations, but the actual effects would need to be measured. The group also removed unnecessary information about classmates, because it had not been given permission to share personal details.
During practice, both teams answered questions from their teacher. Team A explained which numbers were invented and which facts came from a government page. Team B described the original draft’s problem and how it had checked its claims. The teacher said that both groups had learned something useful: tools can offer ideas, but the meaning and evidence of a final presentation must be checked by people.
Finally, students wrote their own reflections. Some wanted to learn how digital tools could assist research, while others preferred writing an outline without them. The teacher did not require everyone to use the same tool. She reminded the class that a suitable method depends on the task, the student’s understanding, school policy, and privacy concerns. Technology can support a careful learner, but it does not remove the need to think.
- Q13. What was the topic of the class presentations?
- Q14. What determined whether generative AI could be used?
- Q15. Where did Team A check its definition of food loss?
- Q16. Were Team A’s practice numbers real data?
- Q17. What unsupported claim did Team B receive?
- Q18. Why could Team B not keep the claim as a fact?
- Q19. How did Team B change the wording?
- Q20. Why did Team B remove information about classmates?
- Q21. Did the teacher say only Team A had learned something useful?
- Q22. What did the teacher say remained a human responsibility?
- Q23. Did all students have to use an AI tool?
- Q24. Summarize the passage in two English sentences.
資料確認・不定詞・英作文8問（Q25〜Q32）
- Q25. Choose: (A) An AI answer is always accurate. (B) An AI answer may need fact-checking.
- Q26. Before using a tool in class, what must students check?
- Q27. Write a polite English request asking where a statistic came from.
- Q28. Change to ask＋人＋to: The teacher / the team / check sources.
- Q29. Change with too … to: The claim was unreliable. We could not use it.
- Q30. Translate: 『彼らは出典を確認しなければなりません』をhave toで。
- Q31. Write a sentence with although showing a benefit and a limitation.
- Q32. Write a two-sentence opinion about responsible use of tools in school.
32問の解答・全根拠
各問1点で32点満点とし、英文要約は意味と文法が自然な別解も認めます。学校が公式に指定する配点とは異なる自習用の採点です。答えが合っていても本文に根拠を示せない場合は再確認しましょう。
Read1の解答Q1〜Q12
- Q1. An English guide about local places. — The first paragraph states the project.
- Q2. Their teacher. — The teacher provided the rules.
- Q3. Suggesting general research questions under supervision. — The permission was limited and conditional.
- Q4. The fictional museum was open on Mondays. — The classroom’s official schedule said it was closed.
- Q5. They checked the official schedule and recorded the source. — Official data was used for verification.
- Q6. Two historical claims lacked reliable sources. — Fluent writing alone did not justify the claims.
- Q7. Information about classmates. — Privacy rules prohibited unnecessary sharing.
- Q8. No. — Students rewrote and verified their final text.
- Q9. Think of general research questions. — The tool’s permitted contribution was brainstorming.
- Q10. The students. — The teacher emphasized human responsibility.
- Q11. The paragraph sounded natural but contradicted the official museum schedule. — Form and accuracy differ.
- Q12. Students used a tool to suggest questions but found an unsupported museum claim. They checked official sources and wrote their own English guide. — Keep suggestion, verification, and responsibility.
Read2の解答Q13〜Q24
- Q13. Food loss. — The first paragraph names the topic.
- Q14. The school’s rules and the teacher’s specific permission. — The class did not have unlimited permission.
- Q15. On a government information page. — The source of the definition was identified.
- Q16. No. They were fictional and labeled as such. — They did not claim to have measured those figures.
- Q17. Its plan would halve food loss at every school. — No evidence supported this universal promise.
- Q18. No one had measured the claimed effect. — A persuasive style is not evidence.
- Q19. It described a possible helpful action and said effects required measurement. — It replaced certainty with an appropriate qualification.
- Q20. It was unnecessary private information. — Privacy was part of responsible editing.
- Q21. No. Both teams learned useful lessons. — The teacher explicitly referred to both.
- Q22. Checking the meaning and evidence of the final work. — The teacher emphasized review.
- Q23. No. — The teacher did not require identical tools for every student.
- Q24. Two teams prepared food-loss presentations and checked the reliability of their information. One revised an unsupported AI-generated claim, and both learned that human judgment and school rules matter. — Key elements: project, verification, and responsibility.
文法・情報活用Q25〜Q32の解答
- Q25. (B) An AI answer may need fact-checking. — Generated responses can include unsupported statements.
- Q26. School rules and the teacher’s instructions. — The articles explicitly require authorization.
- Q27. Could you tell me where this statistic came from? — Embedded question uses where＋subject＋verb.
- Q28. The teacher asked the team to check sources. — SVO＋to不定詞.
- Q29. The claim was too unreliable for us to use. — Result expressed by too＋adjective＋for＋person＋to.
- Q30. They have to check the sources. — Have to＋verb.
- Q31. Although a tool may suggest ideas, people must verify factual claims. — Concession and human responsibility.
- Q32. Digital tools may help students organize ideas. However, students should follow school rules and check facts before presenting them. — Natural alternatives with benefit and caution accepted.
根拠のない断定を避ける英語
『AIが書いたから必ず正しい』『この方法なら全校で食品ロスが半分になる』といった主張は、検証できない場合、事実として書けません。may／could／mightは可能性を表す助動詞ですが、使えば自動的に根拠が生まれるわけでもありません。必要な出典を確認したうえで、自分の意見・提案・未測定の効果を区別しましょう。
7日間の学習方法
1日目はRead1を解き、発見した誤りとその確認方法を整理します。2日目はask＋人＋toと間接疑問の語順、3日目はRead2を解き、事実・意見・仮説を分類。4日目は個人情報・学校規則を確認し、5日目はQ25〜Q32の英作文。6日目に『学校の学習でツールを利用するなら何に注意するか』を4文の英語で説明し、7日目に32問を再テストします。
よくある質問
学校で生成AIを自由に使ってよいですか？
各学校・教員の指示によります。この教材は許可を与えるものではありません。
AIが作った自然な英文は事実といえますか？
自然な文章でも誤情報が含まれる場合があります。重要な主張には出典確認が必要です。
なぜ個人情報を入力しない方がよいのですか？
他の人の情報の取り扱いにはプライバシー・安全・学校規則が関係します。必要性や許可を確認せず共有してはいけません。
この学校の事例は実際にあったことですか？
いいえ。慶英会がStage2読解と情報活用のために独自に作成した架空の教材です。
関連教材と無料体験
学校のNEW TREASUREのAIと人間についての読解・発表で、事実と提案・推測の区別や不定詞の使い方に悩む場合、慶英会では学校の教材範囲と答案に応じて復習内容を整理します。最大1か月完全無料体験・学習相談はこちら。
最大1ヶ月無料体験授業受付中！（残り枠わずかです！）
最大1ヶ月の「完全無料」指導体験
平均2~3時間のレッスン ＋ 毎日LINEサポート付き
ご入塾前に、慶英会の圧倒的な指導品質をじっくりご体感ください。
無料体験だけで成績が上がる方も続出!!
保護者様の教育相談も無料で承ります。
1.学年 2. 学習目標 3.体験ご希望をお送りください。
⚠️【塾長直接指導の残り枠わずか】
枠が埋まり次第終了となります。お早めにお申し込みください。
参考にした主な公式情報
2026年10月26日現在の公式情報を参照。本ページの学校活動・AIによる文章の例・英文2題・設問32問はすべて慶英会が作成した架空の学習教材で、出版社・学校の公式問題や実在のAIの回答の転載ではありません。