Day 29｜実戦演習⑧｜総合問題
【メイン課題】
以下の英文を 6分〜8分以内 に読んで設問に答えましょう。
英文：
The rise of deepfake technology, which uses artificial intelligence to create realistic but fabricated images, videos, and audio recordings, has raised profound concerns about the future of truth and trust in the digital age. Originally developed for entertainment and creative purposes, deepfake tools have become increasingly sophisticated and accessible, allowing virtually anyone with a computer to create convincing fake content within minutes.
The potential for harm is significant and multifaceted. In politics, deepfake videos of public figures saying things they never actually said could be used to manipulate elections, inflame social tensions, or undermine diplomatic relations. During recent election campaigns in several countries, deepfake videos of candidates making controversial statements went viral on social media before they could be identified as fabrications. In the business world, deepfake audio has been used in fraud schemes, with criminals impersonating company executives to authorize fraudulent wire transfers worth millions of dollars.
Perhaps the most devastating personal impact of deepfakes has been in the realm of non-consensual intimate imagery. Deepfake technology has been used to create explicit content featuring individuals, predominantly women, without their knowledge or consent. A 2023 report found that over 95 percent of deepfake content online was non-consensual intimate material. Victims of this abuse face severe psychological harm, damage to their reputations, and difficulties in having the content removed from the internet.
The challenge of detecting deepfakes is an ongoing technological arms race. While researchers have developed AI tools that can identify deepfakes by analyzing subtle inconsistencies in facial movements, lighting, and audio patterns, the technology used to create deepfakes is constantly improving. Each advance in detection capability is quickly countered by improvements in deepfake generation, making it increasingly difficult to distinguish real content from fabricated material.
Media literacy education is widely regarded as one of the most effective long-term responses to the deepfake threat. Teaching people to critically evaluate digital content, verify sources, and recognize the signs of manipulation can help build societal resilience against misinformation. News organizations are developing verification tools and protocols to authenticate content before publication. Social media platforms are implementing policies to label or remove deepfake content, though enforcement remains inconsistent and challenging.
Legal and regulatory frameworks are struggling to keep pace with the technology. Some jurisdictions have enacted laws specifically targeting deepfakes, particularly non-consensual intimate imagery, but many countries still lack adequate legal protections. The challenge lies in balancing the need to prevent harm with the protection of free expression and the legitimate uses of AI-generated content in art, satire, and education. International cooperation is essential, as deepfake content easily crosses national borders through the internet.
Looking ahead, the most effective approach to managing deepfakes will likely involve a combination of technological solutions, legal protections, platform policies, and public education. Digital authentication systems that can verify the origin and integrity of content, similar to how digital signatures verify the authenticity of documents, may provide a foundation for rebuilding trust in digital media. As AI technology continues to advance, society must find ways to harness its creative potential while minimizing the risks it poses to truth, privacy, and democratic institutions.
設問：
（1）ディープフェイクが悪用されている分野を3つ挙げ、それぞれの具体例を答えなさい。
（2）ディープフェイク検出が困難である理由を説明しなさい。
（3）ディープフェイクへの対策として本文で挙げられているものを4つ答えなさい。
（4）本文の内容と一致するものを全て選びなさい。
a. ディープフェイクは元々エンターテインメント目的で開発された
b. 2023年の報告によるとオンラインのディープフェイクの95%以上が非同意の性的コンテンツである
c. 全ての国でディープフェイクを規制する法律が整備されている
d. デジタル認証システムがメディアへの信頼を再構築する基盤になりうる
（5）この文章全体を80字以内で要約しなさい。
模範解答を見る ▼
【模範解答】
各段落のキーワードメモ：
– 第1段落：ディープフェイクの概要
– 第2段落：政治・ビジネスでの悪用
– 第3段落：非同意性的コンテンツの問題
– 第4段落：検出の困難さ
– 第5段落：メディアリテラシーと対策
– 第6段落：法整備の課題
– 第7段落：結論（複合的アプローチ）
設問の解答：
（1）
– ①政治：選挙期間中に候補者が論争的な発言をするディープフェイク動画がSNSで拡散された。
– ②ビジネス：犯罪者が企業幹部を模倣したディープフェイク音声を使い、数百万ドル規模の不正送金を承認させた。
– ③個人（非同意の性的画像）：本人の知識や同意なく、個人（主に女性）をフィーチャーした露骨なコンテンツが作成されている。
（2）ディープフェイクの検出と作成は技術的な「軍拡競争」の状態にある。検出ツールが顔の動きや照明の微妙な不整合を分析して偽コンテンツを識別しても、作成技術が常に改良されるため、検出能力の向上がすぐにディープフェイク生成の改善によって打ち消される。
（3）AIを使ったディープフェイク検出ツール ②メディアリテラシー教育 ③ソーシャルメディアプラットフォームによるラベリングや削除のポリシー ④法的・規制的枠組みの整備
（4）正解：a、b、d
– a：本文 “Originally developed for entertainment and creative purposes” → 正解
– b：本文 “over 95 percent of deepfake content online was non-consensual intimate material” → 正解
– c：本文 “many countries still lack adequate legal protections” → 全ての国ではない → 不正解
– d：本文 “Digital authentication systems … may provide a foundation for rebuilding trust” → 正解
（5）要約例：ディープフェイク技術は政治操作や詐欺、性的悪用に利用され深刻な社会問題となっており、技術的検出・法整備・メディアリテラシー教育を組み合わせた包括的対策が不可欠である。（79字）
ポイント（速読のコツ㉙）：
– 実戦では 時間が足りなくなったときの「緊急戦略」 を持っておこう。残り時間が1分になったら：①未回答の設問を確認 → ②その設問のキーワードで本文をスキャニング → ③該当箇所だけ読んで解答。全文を読み終えなくても、スキャニングで必要な情報さえ見つければ正解できる。
【練習問題】以下の英文を6分〜8分以内に読み、設問に答えなさい。
英文：
The right to privacy in the digital age has become one of the most contentious legal and ethical issues of our time. Every day, billions of people around the world generate enormous amounts of personal data through their online activities, social media posts, shopping transactions, and location-tracking devices. This data is collected, analyzed, and monetized by technology companies, advertisers, and data brokers, often without individuals fully understanding the extent of the information being gathered about them.
The European Union has taken the lead in digital privacy regulation with the General Data Protection Regulation, which came into effect in 2018. The GDPR grants EU citizens the right to know what data companies hold about them, the right to have their data deleted, and the right to refuse data collection. Companies that violate the GDPR face fines of up to 4 percent of their annual global revenue or 20 million euros, whichever is greater. The regulation has had a significant impact beyond Europe, as many global companies have adopted GDPR-compliant practices for all their users worldwide.
In the United States, privacy regulation remains fragmented, with no comprehensive federal law equivalent to the GDPR. Instead, a patchwork of state-level laws and sector-specific regulations governs data privacy. California’s Consumer Privacy Act, enacted in 2020, is the most comprehensive state-level privacy law, giving residents the right to know what personal data is collected about them and to opt out of its sale. Several other states have passed similar legislation, but the lack of a unified national standard creates confusion for both businesses and consumers.
Surveillance technology presents additional privacy challenges. Facial recognition systems deployed in public spaces can identify and track individuals without their knowledge or consent. Many cities and police departments use these systems for law enforcement purposes, arguing that they help solve crimes and enhance public safety. However, studies have shown that facial recognition algorithms can exhibit significant racial and gender biases, producing higher error rates for women and people with darker skin tones. Several cities, including San Francisco and Boston, have banned government use of facial recognition technology due to these concerns.
The tension between privacy and security will continue to shape policy debates for years to come. Governments argue that access to personal data is necessary for national security, counterterrorism, and law enforcement. Privacy advocates counter that unchecked surveillance threatens civil liberties and can be used to suppress dissent and target vulnerable communities. Finding the right balance requires ongoing dialogue between governments, technology companies, civil society organizations, and the public.
設問：
（1）GDPRが市民に与える権利を3つ挙げなさい。
（2）GDPRに違反した企業が受ける罰則を答えなさい。
（3）顔認識技術の問題点を2つ挙げなさい。
（4）この文章全体を80字以内で要約しなさい。
練習問題の解答を見る ▼
【練習問題 解答】
設問の解答：
（1）①企業が保有する自分のデータを知る権利 ②データを削除してもらう権利 ③データ収集を拒否する権利
（2）年間世界売上高の最大4%または2,000万ユーロのいずれか高い方の罰金。
（3）①本人の知識や同意なく公共空間で個人を識別・追跡できる ②顔認識アルゴリズムに人種やジェンダーのバイアスがあり、女性や肌の色が濃い人のエラー率が高い
（4）要約例：デジタル時代のプライバシー保護はEUのGDPRが先行するが米国は規制が断片的で、顔認識技術のバイアス問題も加わり、安全とプライバシーのバランスが課題である。（76字）