[{"data":1,"prerenderedAt":213},["ShallowReactive",2],{"topic-page:\u002Fit\u002Faws\u002Faws-certified-ai-practitioner\u002Fresponsible-ai\u002Fsafety-transparency-and-human-centered-design":3},{"topic":4,"category":169,"seo":173,"breadcrumbs":178,"prerequisites":188,"nextTopics":189,"relatedTopics":201},{"id":5,"slug":6,"title":7,"summary":8,"canonicalPath":9,"categoryPath":10,"difficulty":11,"targetAudience":12,"estimatedMinutes":13,"generatedAt":14,"publishedAt":14,"updatedAt":14,"reviewStatus":15,"tags":16,"expectedKnowledge":26,"learningGoal":29,"keyTerms":30,"blocks":81},"it-aws-aws-certified-ai-practitioner-responsible-ai-safety-transparency-and-human-centered-design","safety-transparency-and-human-centered-design","安全性・透明性のトレードオフと人間中心設計","安全性・透明性のトレードオフと人間中心設計を、AIF-C01の公式objective 4.2.3・4.2.4に沿って、用語、判断順序、例、誤解修正で整理します。","\u002Fit\u002Faws\u002Faws-certified-ai-practitioner\u002Fresponsible-ai\u002Fsafety-transparency-and-human-centered-design","\u002Fit\u002Faws\u002Faws-certified-ai-practitioner\u002Fresponsible-ai","beginner","cert_candidate",6,"2026-08-06","ai_generated",[17,18,19,20,21,22,23,24,25],"AWS Certified AI Practitioner","AIF-C01","責任あるAI","安全性","透明性","解釈可能性","人間中心設計","利用者フィードバック","異議申立て",[27,28],"AWSクラウドの基本用語を聞いたことがある","AIを使う業務例を日常的な言葉で考えられる","安全性・透明性のトレードオフと人間中心設計について、試験で問われる違いと選択基準を説明し、短い業務例へ適用できる。",[31,36,39,42,46,49,53,56,62,66,70,77],{"id":32,"label":33,"shortDefinition":34,"role":35},"human-centered-explainable-ai","人間中心の説明可能AI","利用者の知識・目的・影響に合わせ、理解と適切な行動を支える説明を設計する考え方です。","primary",{"id":37,"label":23,"shortDefinition":38,"role":35},"human-centered-design","利用者の必要・能力・状況を基に、反復的に製品や手順を設計する方法です。",{"id":40,"label":24,"shortDefinition":41,"role":35},"user-feedback","利用者が誤り・有用性・困難を伝え、改善や救済へ戻す情報です。",{"id":43,"label":44,"shortDefinition":45,"role":35},"ai-decision-transparency","AI判断の透明性","AIが使われた事実、根拠、限界、関与範囲を利用者へ明らかにすることです。",{"id":47,"label":25,"shortDefinition":48,"role":35},"appeal-process","利用者がAIの結果へ異議を示し、人による再確認や修正を求められる手続きです。",{"id":50,"label":20,"shortDefinition":51,"role":52},"safety","AIの利用による身体的・心理的・社会的な害を予防し、抑える性質です。","contextual",{"id":54,"label":21,"shortDefinition":55,"role":52},"transparency","AIの目的・所有者・データ・制約・評価・版などを適切に開示できる性質です。",{"id":57,"label":22,"shortDefinition":58,"supplement":59,"role":52},"interpretability","モデルの内部表現や判断過程を、人が理解できる程度です。",{"kind":60,"text":61},"comparison","説明可能性は、判断理由を外部へ示せる性質です。",{"id":63,"label":64,"shortDefinition":65,"role":52},"uncertainty","不確実性","結果が事前に確定せず、予測や生成に幅や誤りが残る性質です。",{"id":67,"label":68,"shortDefinition":69,"role":52},"performance","性能","モデルやシステムが定めた品質・速度・安定性の基準を満たす程度です。",{"id":71,"label":72,"shortDefinition":73,"supplement":74,"role":52},"human-oversight","人による監督","AIの判断や出力を人が確認し、必要に応じて介入・承認する統制です。",{"kind":75,"text":76},"misconception","確認者へ根拠・時間・権限がなければ機能しません。",{"id":78,"label":79,"shortDefinition":80,"role":52},"evaluation","評価","定めたデータと基準で、モデルやアプリの品質・安全性・価値を測る活動です。",[82,87,98,107,116,122,128,134,150,159],{"id":83,"type":84,"title":7,"subtitle":85,"shortDefinition":86},"hero","HeroBlock","性能だけでなく、理解・行動・異議申立てまで設計する","説明を受ける人の知識、目的、時間、影響を考慮し、理解と適切な行動を支えるようAIの表示・操作・救済を設計する考え方です。",{"id":88,"type":89,"term":33,"definition":90,"plainExplanation":91,"presentation":92,"supportLabel":93,"keywords":94},"definition","DefinitionBlock","[[cloze:human-centered-explainable-ai|人間中心の説明可能AI]]とは、利用者の知識・目的・影響に合わせ、理解と適切な行動を支える説明を設計する考え方です。","[[term:human-centered-design|人間中心設計]]で、[[term:ai-decision-transparency|AI判断の透明性]]、根拠、修正方法を相手に合わせます。[[term:user-feedback|利用者フィードバック]]と[[term:appeal-process|異議申立て]]を改善と救済へつなぎます。","natural","言葉を分けると",[95,96,97],"理解","選び直し","救済",{"id":99,"type":100,"steps":101,"title":106},"decision-steps","StepBlock",[102,103,104,105],"利用者が説明後に行うべき判断や行動を定義する。","必要な根拠・[[term:uncertainty|不確実性]]・制約を、理解可能な粒度で示す。","秘密情報や攻撃に使える詳細を公開範囲から除く。","理解度、誤信頼、フィードバック利用、異議処理を実利用者で[[term:evaluation|評価]]する。","説明と安全を両立する順序",{"id":108,"type":109,"title":110,"points":111,"body":115},"exam-cues","KeyPointBlock","判断で押さえる境界",[112,113,114],"公開範囲と安全性を両立","説明後の利用者行動を設計","異議・修正・人への移行を用意","[[term:transparency|透明性]]を高めても、攻撃に使える詳細は[[term:safety|安全性]]のため公開しません。[[term:interpretability|解釈可能性]]と[[term:performance|性能]]の均衡を用途で決め、必要な説明を段階表示します。",{"id":117,"type":118,"src":119,"alt":120,"caption":121},"diagram","DiagramBlock","\u002Fassets\u002Fit\u002Faws\u002Faws-certified-ai-practitioner\u002Fresponsible-ai\u002Fsafety-transparency-and-human-centered-design\u002Fsafety-transparency-and-human-centered-design.svg","安全性・透明性のトレードオフと人間中心設計について理解に必要な説明、不確実性・根拠、秘密・攻撃面の保護、人の選択・救済の関係を文字ラベルと矢印で示した図","説明の便益と、公開し過ぎによる安全リスクを均衡させる図です。中心には利用者の選択と救済を置きます。",{"id":123,"type":124,"scenario":125,"explanation":126,"result":127},"worked-example","ExampleBlock","不正検知システムが取引を保留した理由を顧客へ説明します。","『AIが不正と判断した』だけで終わらず、顧客が確認できる一般的理由、誤判定の可能性、本人確認と異議申立ての方法を示します。検知閾値の詳細は公開せず、[[term:human-oversight|人による監督]]の審査へ進めるようにします。","安全制御を守りながら、顧客が誤りを訂正できる説明になります。",{"id":75,"type":129,"warningTitle":130,"message":131,"severity":132,"action":133},"WarningBlock","人を置くだけでは人間中心にならない","担当者がAI提案をそのまま追認し、根拠や覆す権限がなければ実質的な確認になりません。時間、情報、権限、責任を与えます。","warning","利用者と確認者が、説明を見て何を選び直せるかを具体化します。",{"id":135,"type":136,"title":137,"questions":138},"quiz","QuizBlock","AIF-C01 判断チェック",[139],{"question":140,"choices":141,"answerIndex":145,"choiceExplanations":146},"人間中心の説明として最も適切なのはどれですか。",[142,143,144],"根拠・不確実性・異議申立て方法を相手に分かる形で示す","内部ログを説明なしで全量表示する","AIの判断を変更不能にする",0,[147,148,149],"正解です。理解と救済を支えます。","情報量が多く、秘密や攻撃面も露出します。","人の選択と異議の経路がありません。",{"id":151,"type":152,"title":153,"points":154},"summary","SummaryBlock","まとめ",[155,156,157,158],"透明性と安全性には公開範囲の均衡がある","性能と解釈可能性も用途で比較","説明は利用者の理解と行動を支える","フィードバック・異議・人への移行を設計する",{"id":160,"type":161,"title":162,"nextTopics":163,"relatedKeywords":165},"next-action","NextActionBlock","次は「AWSのAIセキュリティ制御と責任共有」へ",[164],"it-aws-aws-certified-ai-practitioner-ai-security-compliance-governance-aws-security-controls-and-shared-responsibility",[166,167,168],"責任共有モデル","IAM","暗号化",{"path":10,"title":19,"description":170,"parentPath":171,"accentToken":172},"公平性、安全性、データ品質、説明可能性、人間中心設計を、責任あるAIの判断軸として整理します。","\u002Fit\u002Faws\u002Faws-certified-ai-practitioner","cloud-amber",{"title":174,"description":175,"canonicalUrl":176,"ogImage":177},"安全性・透明性のトレードオフと人間中心設計 | AWS Certified AI Practitioner","AIF-C01対策として安全性・透明性のトレードオフと人間中心設計の定義、判断軸、具体例、誤解しやすい境界を公式試験範囲に沿って学びます。","https:\u002F\u002Fzeqnilo.com\u002Fit\u002Faws\u002Faws-certified-ai-practitioner\u002Fresponsible-ai\u002Fsafety-transparency-and-human-centered-design","https:\u002F\u002Fzeqnilo.com\u002Fassets\u002Fsite\u002Fdefault-og.svg",[179,182,185,186,187],{"title":180,"path":181},"IT","\u002Fit",{"title":183,"path":184},"AWS","\u002Fit\u002Faws",{"title":17,"path":171},{"title":19,"path":10},{"title":7,"path":9},[],[190],{"id":164,"title":191,"summary":192,"canonicalPath":193,"categoryPath":194,"categoryTitle":195,"difficulty":11,"targetAudience":12,"estimatedMinutes":13,"publishedAt":14,"updatedAt":14,"reviewStatus":15,"tags":196},"AWSのAIセキュリティ制御と責任共有","AWSのAIセキュリティ制御と責任共有を、AIF-C01の公式objective 5.1.1に沿って、用語、判断順序、例、誤解修正で整理します。","\u002Fit\u002Faws\u002Faws-certified-ai-practitioner\u002Fai-security-compliance-governance\u002Faws-security-controls-and-shared-responsibility","\u002Fit\u002Faws\u002Faws-certified-ai-practitioner\u002Fai-security-compliance-governance","AIのセキュリティ・コンプライアンス・ガバナンス",[17,18,195,166,167,168,197,198,199,200],"AWS PrivateLink","Amazon Macie","Bedrock AgentCore","Guardrails",[202],{"id":203,"title":204,"summary":205,"canonicalPath":206,"categoryPath":10,"categoryTitle":19,"difficulty":11,"targetAudience":12,"estimatedMinutes":13,"publishedAt":14,"updatedAt":14,"reviewStatus":15,"tags":207},"it-aws-aws-certified-ai-practitioner-responsible-ai-transparency-explainability-and-model-tools","透明性・説明可能性とモデル関連ツール","透明性・説明可能性とモデル関連ツールを、AIF-C01の公式objective 4.2.1・4.2.2に沿って、用語、判断順序、例、誤解修正で整理します。","\u002Fit\u002Faws\u002Faws-certified-ai-practitioner\u002Fresponsible-ai\u002Ftransparency-explainability-and-model-tools",[17,18,19,21,208,209,210,211,212],"説明可能性","SageMaker Model Cards","SageMaker Clarify","Bedrock Model Evaluations","ライセンス",1787370540282]