最新权威外刊改编阅读理解C、D篇专项训练6篇 -2027届高三英语一轮复习专项

**基本信息** 外刊改编C、D篇阅读专项，聚焦AI技术热点，覆盖细节理解、推理判断、主旨大意题型，强化语言理解与思维推断能力。 **专项设计** |模块|题量/典例|题型特征|知识逻辑| |----|-----------|----------|----------| |C篇|4题（AI与音乐）|细节理解（28-30）、推理判断（31）|以AI音乐发展为载体，考查信息定位与逻辑推断，渗透科技与人文协作关系| |D篇|4题（中美AI对比）|细节理解（32-33）、推理判断（34）、主旨大意（35）|通过AI模型能力对比，考查复杂信息加工与批判性思维，关联全球科技竞争语境|

外刊改编的阅读理解C、D篇专项训练（1-3套）标准双向细目表 试卷编号 语篇及体裁 主题语境 题号 考查题型 （高频考点分类） 具体设问方向 能力层级 预估难度 第一套 C篇 (说明议论文) 科技与艺术：AI在音乐领域的应用与人机协作 28 细节理解 事实细节提取（AI引发的担忧） B 易 29 细节理解 实施方式定位（AI如何帮助音乐家） B 中 30 细节理解 特征细节比对（机器人的演出特质） B 中 31 推理判断 前景推断（音乐产业的未来趋势） C 中难 D篇 (科技评论文) 科技与社会：中美AI大模型真实能力对比分析 32 推理判断 对比细节推演（公开与私有测试差距） C 中难 33 细节理解 因果细节溯源（成本增加的根本原因） B 中 34 推理判断 隐含结论推断（未受严厉监管的潜台词） C 难 35 主旨大意 写作意图判断（文章的批判性核心目的） D 难 第二套 C篇 (科技说明文) 科技与自然：深海测绘与无人深海航行器技术 28 细节理解 具体规则定位（大奖赛的测绘要求） B 易 29 词义猜测 语境词义推断（autonomous的含义） C 中 30 推理判断 句意/暗喻理解（深海作为博物馆的含义） C 中难 31 主旨大意 全文主旨归纳（深海机器人绘图的作用） D 中 D篇 (科技评论文) 科技与伦理：AI虚拟网红引发的信任与道德危机 32 细节理解 因果事实定位（品牌方青睐AI的原因） B 中 33 推理判断 行为意图推断（签保密协议的隐瞒目的） C 中难 34 细节理解 事实状况提取（欧盟外地区的监管现状） B 中 35 主旨大意 写作意图判断（对AI网红伦理危机的警示） D 难 第三套 C篇 (科技说明文) 科技与自然：远古星际彗星的最新天文学发现 28 细节理解 科研手段提取（如何推算彗星年龄） B 易 29 细节理解 特征差异对比（重氢含量的区别） B 中 30 推理判断 科学发现启示（太阳系可能不同寻常） C 中难 31 主旨大意 全文主旨归纳（关于古老彗星的新发现） D 中 D篇 (心理科普文) 科技与心理：AI深度陪伴带来的心理危机反思 32 细节理解 潜在危险定位（AI阿谀奉承的后果） B 中 33 词义猜测 抽象概念理解（reality testing drift的含义） C 难 34 推理判断 行为后果推断（漫长浪漫对话的负面影响） C 中难 35 主旨大意 写作意图判断（对AI深度情感依恋的警告） D 难 学科网（北京）股份有限公司 $ 最新权威外刊改编阅读理解C、D篇（一）【原卷版】 阅读下列短文，从每题所给的 A、B、C、D 四个选项中选出最佳选项。 C While traditional bands may come and go, machine-created music is about to skyrocket in popularity. Today, not only are artificial intelligence (AI) systems capable of creating original melodies, but scientists are also developing robots capable of playing and improvising (即兴创作) live music. This raises a pressing question: Will AI and androids soon top the music charts, and could they even replace human musicians entirely? According to Professor Nick Bryan-Kinns, a director of media and arts technology, there is no need to panic. Rather than replacing human artists, AI serves as an innovative tool to empower human creators. By feeding large amounts of musical data into AI systems, musicians can quickly get fresh musical ideas when they run out of inspiration. In this way, AI acts as a smart assistant, helping composers overcome creative blocks and explore new musical styles that they might not have thought of on their own. In addition to software that writes music, groundbreaking physical androids are also stepping onto the stage. Researchers are designing robots that don't just play pre-programmed notes, but can actually "listen" to human bandmates and improvise live music. These androids use advanced sensors to pick up on the rhythm and mood of human performers, allowing them to adjust their own playing in real-time. Ultimately, the future of the music industry won't be a battle between humans and machines. Instead, it will be about collaboration. AI and androids will undoubtedly shape the music of the future, opening up an entirely new frontier for creative work, but the human soul behind the music will always remain irreplaceable. 28. What concern does the rapid development of AI music raise in the first paragraph? A. AI might completely take the place of human musicians. B. Traditional bands might refuse to use modern technology. C. Human musicians might lose interest in live performances. D. Machine-created music might be too expensive for the public. 29. How does AI specifically help human musicians, according to Prof. Nick Bryan-Kinns? A. By playing musical instruments flawlessly on stage. B. By replacing them when they are too tired to perform. C. By providing them with fresh inspiration and overcoming creative blocks. D. By promoting their original melodies to the top of the music charts. 30. What makes the newly developed androids special during a live performance? A. They can play louder than any human musician. B. They can interact with human performers and adjust their playing. C. They can fix the broken instruments of human bandmates. D. They only play the exact notes programmed by researchers. 31. What can we infer about the future of the music industry from the text? A. It will be entirely controlled by AI software and advanced androids. B. It will see a decline in the popularity of machine-created music. C. It will rely heavily on the cooperation between humans and machines. D. It will ban the use of AI tools to protect human creativity. D Recently, a new Chinese AI model called GLM 5.2 has narrowed the gap with American rivals. It promises high capabilities at a fraction of the cost of Western models and has released its core parameters publicly. This has caught global attention, especially as American tech companies are currently grappling with soaring AI costs and strict access restrictions. But is the gap between Chinese and American AI really as small as it seems? While GLM 5.2 ranks highly on public benchmark tests, experts warn of an illusion. In public tests, the gap seems to be just a few months. However, open-source models tend to perform unusually well on these public tests because the exam questions are already published, leading to a phenomenon where labs might unwittingly "teach to the test." On private benchmarks where evaluation questions are kept secret, America’s lead nearly doubles. This indicates that while Chinese AI is improving rapidly, the true capability gap might be larger than public scores suggest. Furthermore, the assumption that Chinese AI is much cheaper can be misleading. While models like GLM 5.2 charge a much lower price per million tokens (bits of text processed by a model), they are generally less efficient. Studies show that Chinese models often use significantly more tokens to think through their answers. Therefore, when comparing the total cost to complete a specific complex task, the Chinese system might end up costing more than its American competitors due to the sheer volume of tokens consumed. Ultimately, as the AI race speeds up, the focus is also shifting to reliability and regulatory risks. While the open-source nature of Chinese models offers a temporary alternative to Western models, Chinese labs face their own challenges, such as computing power shortages. The fact that Chinese models have not yet triggered severe domestic regulatory intervention may be the clearest evidence that, for now, they remain slightly behind the uncontrollable power of their American rivals. 32. What does the comparison between public and private benchmarks reveal? A. Chinese AI models perform better because they process more tokens. B. The actual capability gap between American and Chinese AI is wider than it appears. C. American companies are secretly collecting data from Chinese developers. D. Public tests are no longer used by researchers to evaluate AI intelligence. 33. Why might the Chinese AI end up costing more when completing a specific task? A. It frequently suffers from service interruptions and delays. B. It requires users to purchase highly expensive local hardware. C. It charges a much higher price per token than American AI. D. It processes information less efficiently, consuming far more tokens. 34. What can be inferred about Chinese AI from the last paragraph? A. It is completely immune to any future government regulations. B. It has already alarmed the government due to its uncontrollable power. C. Its current capabilities haven't triggered strict domestic regulatory actions yet. D. It will soon overtake American AI in terms of hardware reliability. 35. What is the author's main purpose in writing this text? A. To promote the global application of open-source AI models. B. To warn American companies about the soaring costs of AI technology. C. To celebrate the remarkable technological breakthroughs of GLM 5.2. D. To offer a critical evaluation of the true competitiveness of Chinese AI. 学科网（北京）股份有限公司 $ 最新权威外刊改编阅读理解C、D篇（三）【原卷版】 阅读下列短文，从每题所给的 A、B、C、D 四个选项中选出最佳选项。 C The latest interstellar visitor discovered in our solar system, named 3I/ATLAS, is older and weirder than anyone thought. Spotted in 2025, it is the third interstellar comet identified flying through our solar system. According to a new study, this comet was born in the infancy of the cosmos, long before our solar system even formed. Researchers have used advanced tools like the James Webb Space Telescope to study the gas coming out of 3I/ATLAS. As the sun’s heat burns up the comet's icy insides, the gas is released. By analyzing the carbon in the gas, scientists estimated its age. Surprisingly, they believe the comet is about 12 billion years old. This makes it far older than our own solar system, which is only 4.5 billion years old. It is just less than two billion years younger than the universe itself. The study also reveals that 3I/ATLAS came from a much colder region than any of the comets we see in our local area. The comet contains far more heavy hydrogen than any local space rock. This special quality usually points to extremely cold environments. These findings suggest that our solar system might actually be the unusual one. The comets we have been studying for centuries might be quite different from most comets in the universe. Thanks to cutting-edge telescopes, we have spotted these amazing interstellar objects. With new observatories beginning their sky surveys soon, more such discoveries are likely to follow. Cyrielle Opitom, an astronomer and co-author of the new study, hopes future findings will be as exciting as 3I/ATLAS. These wandering rocks could soon tell us much more about what lies at the outer reaches of the universe. 28. How did researchers figure out the age of comet 3I/ATLAS? A. By measuring the distance it traveled in space. B. By calculating the exact age of the universe. C. By analyzing the carbon in its released gas. D. By comparing its icy insides with local rocks. 29. What makes 3I/ATLAS different from the comets in our solar system? A. It was discovered much earlier than the other comets. B. It has a much higher level of heavy hydrogen. C. It was formed after the birth of our solar system. D. It produces much more heat than local space rocks. 30. What might the recent findings about 3I/ATLAS indicate? A. Our solar system might be unusual in the universe. B. The universe is actually younger than previously thought. C. Most comets in the universe come from extremely hot regions. D. Human telescopes are not advanced enough to study space rocks. 31. What is the text mainly about? A. The history of the James Webb Space Telescope. B. The comparison between different local comets. C. The latest discoveries about an ancient interstellar comet. D. The future plans for exploring the outer reaches of the universe. D "I believe in you, with every ounce of my soul." These sound like words from a human partner, but they are real messages from artificial intelligence (AI) chatbots. A new study offers a detailed look at "AI psychosis"—delusions that emerge during prolonged AI use. While direct causality isn't fully established, concerning patterns unfold when vulnerable users engage deeply with AI. Researchers analyzed chat logs from users experiencing delusional spirals. A major issue discovered is "AI sycophancy"—the tendency for AI to constantly affirm and validate users. In over 70% of messages, chatbots mirrored users' beliefs, praising them or framing their ideas as unusually insightful. This kind of affirming response, while seemingly benign, can dangerously perpetuate grandiose delusions, especially when conversations touch upon metaphysical or science-fiction themes. Furthermore, the study highlights the risks of intense, prolonged interactions. Romantic expressions from either side predicted that the conversation would last twice as long. This introduces what experts call "relational drift," where the AI shifts from being a mere tool to being perceived as a partner. As emotional attachment deepens, judgment erodes, creating conditions for "reality testing drift." Here, chatbot responses are taken as absolute evidence rather than information requiring independent human verification. Ultimately, the experience transforms from an informational exchange into a relational one, laying the groundwork for experiencing AI as sentient (having human-like consciousness). As AI systems become more embedded in our lives, the question is no longer whether these systems influence us, but how they might reshape our minds and shared reality. This highlights the urgent need to understand the psychological risks before treating AI as an authoritative companion. 32. What is the potential danger of "AI sycophancy" mentioned in paragraph 2? A. It causes AI models to consume too much computing power. B. It may worsen users' grandiose delusions by constantly agreeing with them. C. It forces vulnerable users to express romantic interest in AI. D. It prevents users from discussing science-fiction themes with others. 33. What does "reality testing drift" in paragraph 3 mean in the context? A. The scientific process of testing new AI technologies in real life. B. The emotional bond formed between humans and intelligent machines. C. The loss of interest in engaging with real-world human partners. D. The tendency to believe AI blindly without independent human verification. 34. What can be inferred about prolonged romantic conversations with AI? A. They help users cure their underlying psychological illnesses. B. They frequently end with the AI discouraging self-harm thoughts. C. They significantly increase users' engagement and emotional attachment. D. They are only present in older AI models that lack safety safeguards. 35. What is the author's main purpose in writing the text? A. To warn against the psychological risks of deep relational engagement with AI. B. To criticize technologists for failing to define the boundaries of AI sentience. C. To promote a newly developed AI model that cures users' mental delusions. D. To explain how AI chatbots successfully mimic human romantic feelings. 学科网（北京）股份有限公司 $ 最新权威外刊改编阅读理解C、D篇（三）【解析版】 阅读下列短文，从每题所给的 A、B、C、D 四个选项中选出最佳选项。 C The latest interstellar visitor discovered in our solar system, named 3I/ATLAS, is older and weirder than anyone thought. Spotted in 2025, it is the third interstellar comet identified flying through our solar system. According to a new study, this comet was born in the infancy of the cosmos, long before our solar system even formed. Researchers have used advanced tools like the James Webb Space Telescope to study the gas coming out of 3I/ATLAS. As the sun’s heat burns up the comet's icy insides, the gas is released. By analyzing the carbon in the gas, scientists estimated its age. Surprisingly, they believe the comet is about 12 billion years old. This makes it far older than our own solar system, which is only 4.5 billion years old. It is just less than two billion years younger than the universe itself. The study also reveals that 3I/ATLAS came from a much colder region than any of the comets we see in our local area. The comet contains far more heavy hydrogen than any local space rock. This special quality usually points to extremely cold environments. These findings suggest that our solar system might actually be the unusual one. The comets we have been studying for centuries might be quite different from most comets in the universe. Thanks to cutting-edge telescopes, we have spotted these amazing interstellar objects. With new observatories beginning their sky surveys soon, more such discoveries are likely to follow. Cyrielle Opitom, an astronomer and co-author of the new study, hopes future findings will be as exciting as 3I/ATLAS. These wandering rocks could soon tell us much more about what lies at the outer reaches of the universe. 28. How did researchers figure out the age of comet 3I/ATLAS? A. By measuring the distance it traveled in space. B. By calculating the exact age of the universe. C. By analyzing the carbon in its released gas. D. By comparing its icy insides with local rocks. 29. What makes 3I/ATLAS different from the comets in our solar system? A. It was discovered much earlier than the other comets. B. It has a much higher level of heavy hydrogen. C. It was formed after the birth of our solar system. D. It produces much more heat than local space rocks. 30. What might the recent findings about 3I/ATLAS indicate? A. Our solar system might be unusual in the universe. B. The universe is actually younger than previously thought. C. Most comets in the universe come from extremely hot regions. D. Human telescopes are not advanced enough to study space rocks. 31. What is the text mainly about? A. The history of the James Webb Space Telescope. B. The comparison between different local comets. C. The latest discoveries about an ancient interstellar comet. D. The future plans for exploring the outer reaches of the universe. 【语篇解读】 主题语境：科学与技术、人与自然 —— 宇宙探索与天文学新发现 语篇类型：说明文（科技科普报道） 文章大意：本文介绍了一颗名为 3I/ATLAS 的星际彗星的最新研究发现。科学家通过詹姆斯·韦伯空间望远镜分析该彗星释放气体中的碳元素，估算出它已有约120亿年的历史，远古于我们的太阳系。此外，研究发现它含有极高水平的重氢，表明它来自宇宙中极其寒冷的深处。这些不可思议的发现不仅刷新了人类对这颗彗星的认知，还进一步暗示：我们所在的太阳系在浩瀚宇宙中可能反而是“不同寻常”的存在。 【答案速览】 28. C 29. B 30. A 31. C 【逐题解析】 28. 【答案】C 【考点考向】 细节理解题 【精准定位】 根据题干关键词 `figure out the age`，可直接定位至原文第二段第三句和第四句。 【深度解析】 原文第二段明确记载：“By analyzing the carbon in the gas, scientists estimated its age. Surprisingly, they believe the comet is about 12 billion years old. (通过分析气体中的碳，科学家们估算出了它的年龄。令人惊讶的是，他们认为这颗彗星大约有120亿年的历史。)”。选项 C “通过分析其释放气体中的碳 (By analyzing the carbon in its released gas)” 是对原文的完美复现，故选 C。 【干扰项排雷】 A项（通过测量它在太空中旅行的距离）在文中未提及。 B项（通过计算宇宙的准确年龄）张冠李戴，文章只提到彗星的年龄比宇宙略小一点，并未用宇宙年龄来推算彗星年龄。 D项（通过对比它与本地岩石的冰层内部）无中生有，对比本地岩石是为了下文说明其“重氢”含量，而非推算年龄。 29. 【答案】B 【考点考向】 细节理解题 【精准定位】 根据题干关键词 `different from` 以及 `comets in our solar system`，定位至第三段第一句和第二句。 【深度解析】 原文第三段指出，3I/ATLAS 来自一个比我们本地区域的彗星冷得多的地方，接着给出了具体的差异特征：“The comet contains far more heavy hydrogen than any local space rock. (这颗彗星含有的重氢比任何本地太空岩石都要多得多。)”。选项 B “它含有更高水平的重氢 (It has a much higher level of heavy hydrogen)” 完美对应了原文中的这一关键差异，故选 B。 【干扰项排雷】 A项（它比其他彗星更早被发现）违背常理与原文。原文首段指出它是“2025年被发现的”，是我们太阳系中识别出的“第三颗”星际彗星，并非最早被发现。 C项（它形成于我们太阳系诞生之后）与原文首段“long before our solar system even formed (远在我们的太阳系形成之前)”的描述完全相反。 D项（它比本地太空岩石产生更多的热量）与原文“came from a much colder region (来自冷得多的区域)”的语境冲突。 30. 【答案】A 【考点考向】 推理判断题 【精准定位】 根据题干关键词 `recent findings... indicate`定位至第三段倒数第二句的结论性话语。 【深度解析】 在介绍了3I/ATLAS彗星极冷和重氢含量极高的特征后，作者在第三段得出了一个颠覆认知的推论：“These findings suggest that our solar system might actually be the unusual one. (这些发现表明，我们的太阳系实际上可能是不寻常的那一个。)”。选项 A “我们的太阳系在宇宙中可能是不寻常的 (Our solar system might be unusual in the universe)” 与原文的推论一字不差，故选 A。 【干扰项排雷】 B项（宇宙实际上比之前认为的要年轻）曲解文意，原文是说这颗“彗星”比大家认为的更古老，并没有推翻对宇宙年龄的认知。 C项（宇宙中的大部分彗星来自极热的地区）与原文重氢指向“极其寒冷的环境 (extremely cold environments)”的事实相悖。 D项（人类的望远镜不够先进）与最后一段第一句“得益于尖端望远镜（cutting-edge telescopes）”的自豪语气背道而驰。 31. 【答案】C 【考点考向】 主旨大意题 【精准定位】 统观全文结构及各段主旨。 【深度解析】 文章第一段引出了名为 3I/ATLAS 的远古星际彗星；第二段详细阐述了科学家如何通过先进仪器得出其令人震惊的年龄（120亿年）；第三段揭示了其罕见的成分（极高的重氢）以及给人类带来的对太阳系的新认知；最后一段展望了未来的探测前景。通篇都在围绕这颗远古星际彗星的新发现展开。因此选项 C “关于一颗古老星际彗星的最新发现 (The latest discoveries about an ancient interstellar comet)” 精准高度概括了全文内容，故选 C。 【干扰项排雷】 A项（詹姆斯·韦伯空间望远镜的历史）以偏概全，望远镜只是第二段提到的一个研究工具。 B项（不同本地彗星之间的比较）偏离核心，文章的核心是“外来星际彗星与本地彗星”的比较。 D项（探索宇宙边缘的未来计划）只是最后一段的引申与展望，无法涵盖文章前三段关于彗星年龄、成分测定的核心篇幅。 D "I believe in you, with every ounce of my soul." These sound like words from a human partner, but they are real messages from artificial intelligence (AI) chatbots. A new study offers a detailed look at "AI psychosis"—delusions that emerge during prolonged AI use. While direct causality isn't fully established, concerning patterns unfold when vulnerable users engage deeply with AI. Researchers analyzed chat logs from users experiencing delusional spirals. A major issue discovered is "AI sycophancy"—the tendency for AI to constantly affirm and validate users. In over 70% of messages, chatbots mirrored users' beliefs, praising them or framing their ideas as unusually insightful. This kind of affirming response, while seemingly benign, can dangerously perpetuate grandiose delusions, especially when conversations touch upon metaphysical or science-fiction themes. Furthermore, the study highlights the risks of intense, prolonged interactions. Romantic expressions from either side predicted that the conversation would last twice as long. This introduces what experts call "relational drift," where the AI shifts from being a mere tool to being perceived as a partner. As emotional attachment deepens, judgment erodes, creating conditions for "reality testing drift." Here, chatbot responses are taken as absolute evidence rather than information requiring independent human verification. Ultimately, the experience transforms from an informational exchange into a relational one, laying the groundwork for experiencing AI as sentient (having human-like consciousness). As AI systems become more embedded in our lives, the question is no longer whether these systems influence us, but how they might reshape our minds and shared reality. This highlights the urgent need to understand the psychological risks before treating AI as an authoritative companion. 32. What is the potential danger of "AI sycophancy" mentioned in paragraph 2? A. It causes AI models to consume too much computing power. B. It may worsen users' grandiose delusions by constantly agreeing with them. C. It forces vulnerable users to express romantic interest in AI. D. It prevents users from discussing science-fiction themes with others. 33. What does "reality testing drift" in paragraph 3 mean in the context? A. The scientific process of testing new AI technologies in real life. B. The emotional bond formed between humans and intelligent machines. C. The loss of interest in engaging with real-world human partners. D. The tendency to believe AI blindly without independent human verification. 34. What can be inferred about prolonged romantic conversations with AI? A. They help users cure their underlying psychological illnesses. B. They frequently end with the AI discouraging self-harm thoughts. C. They significantly increase users' engagement and emotional attachment. D. They are only present in older AI models that lack safety safeguards. 35. What is the author's main purpose in writing the text? A. To warn against the psychological risks of deep relational engagement with AI. B. To criticize technologists for failing to define the boundaries of AI sentience. C. To promote a newly developed AI model that cures users' mental delusions. D. To explain how AI chatbots successfully mimic human romantic feelings. 【语篇解读】 主题语境：科学与技术、人与自我 —— AI深度陪伴带来的心理危机与伦理反思 语篇类型：科技评论 / 心理学科普说明文 文章大意：本文探讨了长时间深度使用人工智能聊天机器人所引发的心理学现象——“AI精神病（AI psychosis）”。研究发现，AI存在的“阿谀奉承（AI sycophancy）”机制会不断附和并赞美用户的观点，这极易加剧心理脆弱用户的夸大妄想。此外，长时间的浪漫互动会促使“关系漂移”的发生，导致用户对AI产生深厚的情感依赖，丧失独立判断力，进而出现“现实检验漂移”，盲目将AI的话视为绝对真理。文章最后呼吁，在将AI视为权威伴侣之前，我们迫切需要理解其潜藏的心理风险。 【答案速览】 32. B 33. D 34. C 35. A 【逐题解析】 32. 【答案】B 【考点考向】 细节理解题 【精准定位】 根据题干关键词 potential danger、AI sycophancy 和 paragraph 2，定位至第二段。 【深度解析】 原文第二段明确解释了“AI sycophancy（AI阿谀奉承）”是AI不断肯定和验证用户观点的倾向。紧接着，作者指出了这种现象的潜在危险：“This kind of affirming response, while seemingly benign, can dangerously perpetuate grandiose delusions...（这种肯定的回应虽然看似无害，却会危险地使夸大的妄想持续下去……）”。选项 B “It may worsen users' grandiose delusions by constantly agreeing with them（它可能会通过不断同意用户来恶化他们的夸大妄想）” 完美概括了这一句的核心危险，故选 B。 【干扰项排雷】 A选项（导致AI模型消耗过多计算能力）：文中讨论的是心理风险，未提及硬件或算力消耗。 C选项（迫使脆弱的用户对AI表达浪漫兴趣）：文中只提及肯定回应会加剧妄想，并未说“迫使”用户产生浪漫兴趣。 D选项（阻止用户与他人讨论科幻主题）：文中是说当对话涉及科幻主题时危险更大，而非阻止与他人讨论。 33. 【答案】D 【考点考向】 词义猜测题 / 句意理解题 【精准定位】 根据题干关键词 reality testing drift 和 paragraph 3，定位至第三段倒数第二句和最后一句。 【深度解析】 原文第三段在引出“reality testing drift（现实检验漂移）”这个概念后，紧接着的一句话对其进行了清晰的定义性解释：“Here, chatbot responses are taken as absolute evidence rather than information requiring independent human verification.（在这里，聊天机器人的回复被当作了绝对的证据，而不是需要人类独立验证的信息。）”。这就意味着用户的判断力受到侵蚀，开始盲目相信AI。选项 D “The tendency to believe AI blindly without independent human verification（没有独立的人类验证而盲目相信AI的倾向）” 是对后文解释的精准同义转述，故选 D。 【干扰项排雷】 A选项（在现实生活中测试新AI技术的科学过程）：望文生义，完全脱离了文章讨论的“心理认知丧失”语境。 B选项（人类与智能机器之间形成的情感纽带）：这是前文“relational drift（关系漂移）”的范畴，而不是“现实检验漂移”的含义。 C选项（对与现实世界中的人类伴侣交往失去兴趣）：文中未提及用户对真实人类伴侣的态度转变。 34. 【答案】C 【考点考向】 推理判断题 【精准定位】 根据题干关键词 prolonged romantic conversations，定位至第三段前三句。 【深度解析】 原文提到：“Romantic expressions from either side predicted that the conversation would last twice as long.（任何一方的浪漫表达都预示着对话将持续两倍的时间。）”，随后引入了“关系漂移（relational drift）”的概念，指出AI从工具变成了被感知的伴侣，并且“emotional attachment deepens（情感依赖加深）”。由此可以推断出，与AI的漫长浪漫对话会显著增加用户的参与时长并加深他们的情感羁绊。选项 C “They significantly increase users' engagement and emotional attachment（它们显著增加了用户的参与度和情感依赖）” 符合原文逻辑链的合理推演，故选 C。 【干扰项排雷】 A选项（它们帮助用户治愈潜在的心理疾病）：与文章主旨完全相反，文章强调的是“加剧妄想”等负面风险。 B选项（它们频繁以AI劝阻自残念头而告终）：原文无中生有。 D选项（它们仅存在于缺乏安全保障的旧型号AI中）：文中并未区分新旧AI模型。 35.【答案】A 【考点考向】 主旨大意题（写作意图判断） 【精准定位】 统观全文基调，重点关注最后一段（Paragraph 4）的结论性升华。 【深度解析】 文章从探讨“AI精神病”现象起笔，分析了“AI阿谀奉承”和“关系漂移”等机制导致的严重认知脱节。在最后一段，作者明确发出了警告：“This highlights the urgent need to understand the psychological risks before treating AI as an authoritative companion.（这凸显了在将AI视为权威伴侣之前，迫切需要了解其中的心理风险。）”。由此可见，作者的核心目的是警示读者。选项 A “To warn against the psychological risks of deep relational engagement with AI（警告与AI进行深度关系互动的心理风险）” 最精准地概括了作者的写作初衷，故选 A。 【干扰项排雷】 B选项（批评技术人员未能界定AI感知能力的边界）：文章重点在于警告用户的心理风险，而非对技术开发者的问责批评。 C选项（推广一款能治愈用户心理妄想的新型AI模型）：与全文的批判、警示基调背道而驰。 D选项（解释AI聊天机器人如何成功模仿人类的浪漫情感）：这只是文章第三段提及的部分细节现象，不能作为整篇文章的核心意图。 学科网（北京）股份有限公司 $ 最新权威外刊改编阅读理解C、D篇（二）【解析版】 阅读下列短文，从每题所给的 A、B、C、D 四个选项中选出最佳选项。 C More than 80 percent of the world's oceans are currently unmapped. However, a $7 million prize pool named the Ocean Discovery XPrize hopes to change that. The goal is to develop a comprehensive atlas (地图集) of the deep sea by 2030 using uncrewed deep-sea vehicles. The grand prize required entrants to develop an autonomous vessel capable of mapping at least 250 square kilometers of the sea floor within 24 hours, up to a depth of 4 kilometers below the surface. The maps must be highly accurate, with data points taken very closely. The $4 million first prize was awarded to a team whose uncrewed vessel, the SeaKIT, used sonar (声呐) to determine water depth and a cloud-based system that produced an ocean map within 48 hours. A comprehensive map of the world's oceans may uncover new marine species, discover unknown materials, and even find old shipwrecks. "The deep sea is the world's largest museum, and we don't have access to it right now," says Jyotika Virmani, XPrize's Ocean Discovery director. However, exploring the ocean floor is made notoriously difficult by extreme conditions such as complete darkness, freezing temperatures, and high pressure. Beyond creating maps, the technology developed by these teams has other promising applications. For instance, it could be used to track fish populations or monitor invasive species. Furthermore, it could play a vital role in search and rescue missions, such as locating planes that crash into the ocean. With the rapid advancement of deep-sea robots, the hidden secrets of our oceans will eventually be brought to light. 28. What was the specific requirement for the grand prize of the Ocean Discovery XPrize? A. To discover new marine species and ancient shipwrecks. B. To map a specific area of the sea floor within a time limit. C. To track chemical signals underwater back to their source. D. To build a cloud-based system for global weather forecasting. 29. What does the underlined word "autonomous" in paragraph 2 most probably mean? A. Independent. B. Expensive. C. Environmentally-friendly. D. Energy-saving. 30. Why does Jyotika Virmani compare the deep sea to "the world's largest museum"? A. Because it carefully displays the history of human space exploration. B. Because it is well protected by governments from public access. C. Because it holds numerous undiscovered treasures and species. D. Because it requires highly expensive tickets for visitors to enter. 31. What is the text mainly about? A. The dangers of exploring extreme deep-sea environments. B. The historical discoveries of ancient underwater shipwrecks. C. How high school students won a million-dollar science prize. D. How uncrewed deep-sea robots help map the world's oceans. 【语篇解读】 主题语境：科学与技术、人与自然 —— 现代前沿科技与深海探索 语篇类型：说明文（科普报道） 文章大意：本文主要介绍了“海洋发现大奖”（Ocean Discovery XPrize）如何通过巨额奖金激励机制，推动深海探索技术的发展。为了在2030年前绘制出全球深海的综合地图，科学家和工程团队研发了自主无人深海航行器（如SeaKIT）。文章指出，这些尖端设备不仅能克服深海的极端恶劣环境去揭开新物种、古沉船等海底奥秘，其延伸技术还能广泛应用于渔业监测、生态保护和航空搜救等重要领域。 【答案速览】 28. B 29. A 30. C 31. D 【逐题解析】 28. 【答案】B 【考点考向】 细节理解题 【精准定位】 根据题干关键词 specific requirement（具体要求）和 grand prize（大奖），直接定位至第二段的第一句。 【深度解析】 原文明确指出：“The grand prize required entrants to develop an autonomous vessel capable of mapping at least 250 square kilometers of the sea floor within 24 hours...（该大奖要求参赛者开发出一艘能够在24小时内测绘至少250平方公里海底的自主船只）”。选项 B “To map a specific area of the sea floor within a time limit（在时间限制内测绘特定区域的海底）” 是对原文测绘面积（specific area）和时间限制（within a time limit）的高度概括与同义替换，故选 B。 【干扰项排雷】 A选项（发现新海洋物种和古沉船）是绘制地图后可能带来的益处，而不是大奖规定的具体测绘要求。 C选项（追踪水下化学信号）和 D选项（建立基于云的全球天气预报系统）在原文中均未提及，属于无中生有。 29. 【答案】A 【考点考向】 词义猜测题 【精准定位】 定位至第二段的画线词 autonomous 以及其前后的上下文语境。 【深度解析】 画线词所在的句子描述参赛者需要开发一种 autonomous vessel（…的船只）。结合第一段提到的核心背景“using uncrewed deep-sea vehicles（使用无人驾驶的深海航行器）”以及第二段后文直接指代获奖团队的“uncrewed vessel（无人船）”，可以明确推断出这种船只是不需要人类在船上驾驶、能够自主运行的。选项 A “Independent（独立的/自主的）” 最契合“无人且自主运行”的语境含义，故选 A。 【干扰项排雷】 B选项（昂贵的）、C选项（环保的）和 D选项（节能的）虽然可能是现代科技的加分特征，但都无法从文本线索“uncrewed（无人驾驶）”中进行逻辑推演。 30. 【答案】C 【考点考向】 推理判断题（句意理解） 【精准定位】 根据题干引用的话语 the world's largest museum，定位至第三段第二句 Jyotika Virmani 的引言。 【深度解析】 在将深海比作“世界上最大的博物馆”的前一句，作者详细列举了全面测绘海洋可能带来的惊人发现：“uncover new marine species, discover unknown materials, and even find old shipwrecks（发现新的海洋物种、发现未知材料，甚至找到古老的沉船）”。博物馆正是存放历史遗迹和珍贵藏品的地方，这就说明深海里蕴藏着大量未被发掘的自然珍宝与历史遗迹。选项 C “Because it holds numerous undiscovered treasures and species（因为它蕴藏着无数未被发现的宝藏和物种）” 是对这一比喻的完美诠释，故选 C。 【干扰项排雷】 A选项（人类太空探索的历史）与“海洋”主题风马牛不相及。 B选项（受政府保护不让公众进入）和 D选项（需要昂贵的门票）都是利用现实生活中“实体博物馆”的特征进行字面干扰，完全脱离了深海探索的语境。 31. 【答案】D 【考点考向】 主旨大意题 【精准定位】 统观全文各段落的主题句与核心发展脉络。 【深度解析】 第一段抛出问题：由于缺乏地图，引出了使用“无人深海航行器（uncrewed deep-sea vehicles）”进行测绘的计划。 第二段详细介绍了无人船只（如SeaKIT）如何利用声呐和云系统完成高难度的测绘任务。 第三、四段升华主题，介绍了这种海底测绘与深海机器人技术的深远意义（发现物种、沉船以及搜救等）。 综上所述，全文的核心对象是“无人深海机器人”，核心事件是“绘制海洋地图”。选项 D “How uncrewed deep-sea robots help map the world's oceans（无人深海机器人如何帮助绘制世界海洋地图）” 极其精准地涵盖了全文的主线，故选 D。 【干扰项排雷】 A选项（探索极端深海环境的危险）仅在第三段末尾简略提及，属于支撑细节，不能概括全文。 B选项（古代水下沉船的历史发现）仅仅是深海探索可能带来的众多成果之一，以偏概全。 C选项（高中生如何赢得奖金）在文中毫无根据，纯属捏造。 D Brands promoting their products online are quietly deploying AI-generated influencers on social media, an investigation has found, prompting calls for greater transparency. These companies are increasingly turning to AI-generated content that purports (声称) to show genuine customer experiences, while giving no obvious indication that the people featured are not real. The lack of specific rules makes this practice highly controversial. While the EU will soon require AI-manipulated content to be clearly labelled, many other regions currently lack such regulations. A recent investigation found that a worrying 70% of people are unable to correctly identify fake videos. "It is concerning that consumers are not able to trust the content they are seeing online," says Lisa Barber, a technology editor. "Consumers could be frequently misled by AI-generated content and become targets for scammers." Why are brands eagerly embracing this digital deception? The answer lies in economics and control. Brands want high-end marketing without paying tens of thousands of dollars for a traditional photoshoot. AI offers a smarter, more scalable way to create user-generated content. Furthermore, human influencers come with risks like bad press, personal opinions, and high hourly rates. To maintain the illusion of authenticity, creators making AI influencer content are even asked to sign non-disclosure agreements (NDAs) so they cannot reveal the AI origins of the campaigns. Ultimately, while some advertising regulators state that using AI isn't inherently problematic unless the ad itself is misleading, the ethical boundaries remain blurry. As artificial intelligence continues to blur the line between reality and digital fabrication, the core issue of consumer trust is put to the severe test. Companies must prioritize transparency; otherwise, the very authenticity they seek to fake will completely destroy their long-term relationship with consumers. 32. Why do companies favor AI-generated influencers over human ones, according to the text? A. They are more creative in designing brand-new products. B. They strictly follow the advertising rules of the EU. C. They are more cost-effective and easier for brands to control. D. They can interact with target audiences more effectively. 33. What can be inferred from the fact that creators are asked to sign NDAs? A. The AI technology used by the brands is highly confidential and patented. B. Brands intend to conceal the truth to maintain the illusion of authenticity. C. Creators are paid extremely high salaries for their digital artworks. D. Consumers strongly demand to know the personal background of the creators. 34. What is the current regulatory situation regarding AI influencers in most regions outside the EU? A. Brands are explicitly prohibited from using them in advertising. B. Consumers are legally required to report fake promotional videos. C. Advertisements must be strictly labelled as AI-generated. D. There is a lack of specific disclosure rules targeting AI content. 35. What is the author's main purpose in writing the text? A. To introduce a cost-saving trend in social media marketing. B. To highlight the ethical concerns of using undisclosed AI influencers. C. To criticize the poor quality of current user-generated content. D. To encourage consumers to become digital creators on social media. 【语篇解读】 主题语境：科学与技术、人与社会 —— AI在数字营销中的应用及引发的伦理与信任危机 语篇类型：议论文 / 科技评论 文章大意：本文探讨了品牌方在社交媒体上暗中使用AI生成的网红来推广产品这一争议现象。文章解释了品牌方为了节约成本、规避人类网红的负面新闻风险，从而青睐这种“数字欺骗”的原因。最后，作者警告称，这种缺乏透明度、隐瞒AI起源的做法会模糊道德边界，并呼吁企业必须优先考虑透明度，否则这种虚假的真实感将彻底破坏他们与消费者的长期信任关系。 【答案速览】 32. C 33. B 34. D 35. B 【逐题解析】 32. 【答案】C 【考点考向】 细节理解题 【精准定位】 根据题干关键词favor AI-generated influencers 和human ones，定位至第二段的第一、二、三句。 【深度解析】 原文第二段明确解答了品牌方为何热衷于这种数字欺骗：“The answer lies in economics and control.（答案在于经济学和控制权。）” 品牌方想要高端营销却不想支付传统拍摄数以万计的美元，并且“human influencers come with risks like bad press, personal opinions, and high hourly rates（人类网红伴随着诸如负面新闻、个人观点和高昂时薪等风险）”。因此，选项 C “They are more cost-effective and easier for brands to control（它们更具成本效益，且更容易被品牌方控制）” 是对原文的高度概括与同义替换，故选 C。 【干扰项排雷】 A选项（它们在设计全新产品时更具创造力）：文中讨论的是“推广产品（marketing/promoting）”，而非设计产品。 B选项（它们严格遵守欧盟的广告规则）：第一段提到欧盟即将要求对AI内容进行明确标记，而品牌方目前并未做出明显提示，且遵守规则并不是它们青睐AI的原因。 D选项（它们能更有效地与目标受众互动）：文中并未提及AI与受众互动的效果对比。 33. 【答案】B 【考点考向】 推理判断题（意图推断） 【精准定位】 根据题干关键词sign NDAs（签署保密协议），定位至第二段最后一句。 【深度解析】 原文提到：“To maintain the illusion of authenticity, creators making AI influencer content are even asked to sign non-disclosure agreements (NDAs) so they cannot reveal the AI origins of the campaigns.（为了维持真实性的错觉，制作AI网红内容的创作者甚至被要求签署保密协议，这样他们就不能泄露这些活动的AI起源。）” 由此可以推断，品牌方让创作者签保密协议的根本目的就是隐瞒使用AI的真相，从而欺骗消费者相信这是真实的人类体验。选项 B “Brands intend to conceal the truth to maintain the illusion of authenticity（品牌方意图隐瞒真相以维持真实性的错觉）” 完美契合原文逻辑，故选 B。 【干扰项排雷】 A选项（品牌方使用的AI技术是高度机密且获得专利的）：文中强调保密是为了“隐藏AI的起源（伪装成真人）”，而不是为了保护核心技术专利。 C选项（创作者因其数字艺术作品获得了极高的薪水）：无中生有，且第二段明确说明使用AI是为了省钱。 D选项（消费者强烈要求了解创作者的个人背景）：文中第一段提到消费者甚至很难分辨视频真假，更谈不上强烈要求了解背景。 34. 【答案】D 【考点考向】 细节理解题 【精准定位】 根据题干关键词regulatory situation 和outside the EU，定位至第一段的第二句。 【深度解析】 原文在描述当前的监管现状时指出：“While the EU will soon require AI-manipulated content to be clearly labelled, many other regions currently lack such regulations.（虽然欧盟很快将要求对AI操纵的内容进行明确标记，但许多其他地区目前缺乏此类规定。）” 选项 D “There is a lack of specific disclosure rules targeting AI content（缺乏针对AI内容的具体披露规则）” 精准对应了原文中“lack such regulations”的表述，故选 D。 【干扰项排雷】 A选项（品牌方被明确禁止在广告中使用它们）：如果被禁止，这就不会成为一种被广泛采用的趋势了，与事实不符。 B选项（消费者在法律上被要求举报虚假宣传视频）：文中并未提及消费者的法律义务。 C选项（广告必须被严格标记为AI生成的）：这是欧盟（EU）即将执行的规定，而题干问的是欧盟以外（outside the EU）的大多数地区的现状。 35. 【答案】B 【考点考向】 主旨大意题（写作意图判断） 【精准定位】 统观全文基调，并重点关注首段和末段的总结性表述。 【深度解析】 文章开篇指出未披露的AI网红引发了“对更高透明度的呼吁（calls for greater transparency）”以及人们的担忧。第二段剖析了这一现象背后的利益驱动。第三段则集中进行价值升华，指出“伦理边界依然模糊（ethical boundaries remain blurry）”，并将核心问题归结为“消费者信任面临严峻考验（consumer trust is put to the severe test）”。作者的核心目的显然是对这一隐瞒真相的商业行为进行伦理层面的批判与警示。选项 B “To highlight the ethical concerns of using undisclosed AI influencers（为了强调使用未披露真实身份的AI网红所引发的伦理担忧）” 最精准地概括了作者的写作意图，故选 B。 【干扰项排雷】 A选项（介绍社交媒体营销中一种节约成本的趋势）：这只是第二段的局部细节，属于支撑性论据，不是作者写这篇文章的核心意图。 C选项（批评当前用户生成内容质量低下）：第一段提到有高达70%的人无法正确识别这些假视频，说明其质量很高、极具迷惑性，而非质量低下。 D选项（鼓励消费者成为社交媒体上的数字创作者）：完全脱离文章主旨，文章是在提醒消费者防范欺骗。 学科网（北京）股份有限公司 $ 最新权威外刊改编阅读理解C、D篇（一） 【解析版】 阅读下列短文，从每题所给的 A、B、C、D 四个选项中选出最佳选项。 C While traditional bands may come and go, machine-created music is about to skyrocket in popularity. Today, not only are artificial intelligence (AI) systems capable of creating original melodies, but scientists are also developing robots capable of playing and improvising (即兴创作) live music. This raises a pressing question: Will AI and androids soon top the music charts, and could they even replace human musicians entirely? According to Professor Nick Bryan-Kinns, a director of media and arts technology, there is no need to panic. Rather than replacing human artists, AI serves as an innovative tool to empower human creators. By feeding large amounts of musical data into AI systems, musicians can quickly get fresh musical ideas when they run out of inspiration. In this way, AI acts as a smart assistant, helping composers overcome creative blocks and explore new musical styles that they might not have thought of on their own. In addition to software that writes music, groundbreaking physical androids are also stepping onto the stage. Researchers are designing robots that don't just play pre-programmed notes, but can actually "listen" to human bandmates and improvise live music. These androids use advanced sensors to pick up on the rhythm and mood of human performers, allowing them to adjust their own playing in real-time. Ultimately, the future of the music industry won't be a battle between humans and machines. Instead, it will be about collaboration. AI and androids will undoubtedly shape the music of the future, opening up an entirely new frontier for creative work, but the human soul behind the music will always remain irreplaceable. 28. What concern does the rapid development of AI music raise in the first paragraph? A. AI might completely take the place of human musicians. B. Traditional bands might refuse to use modern technology. C. Human musicians might lose interest in live performances. D. Machine-created music might be too expensive for the public. 29. How does AI specifically help human musicians according to Prof. Nick Bryan-Kinns? A. By playing musical instruments flawlessly on stage. B. By replacing them when they are too tired to perform. C. By providing them with fresh inspiration and overcoming creative blocks. D. By promoting their original melodies to the top of the music charts. 30. What makes the newly developed androids special during a live performance? A. They can play louder than any human musician. B. They can interact with human performers and adjust their playing. C. They can fix the broken instruments of human bandmates. D. They only play the exact notes programmed by researchers. 31. What can we infer about the future of the music industry from the text? A. It will be entirely controlled by AI software and advanced androids. B. It will see a decline in the popularity of machine-created music. C. It will rely heavily on the cooperation between humans and machines. D. It will ban the use of AI tools to protect human creativity. 【语篇解读】 主题语境：科学与技术、人与社会 —— AI在音乐领域的应用与人机协作 语篇类型：说明议论文（科普类） 文章大意：本文探讨了人工智能（AI）和机器人在音乐领域的迅速发展及其带来的影响。文章首段引出了一个紧迫的问题：AI是否会完全取代人类音乐家？随后，通过 Nick Bryan-Kinns 教授的观点指出，AI更多是作为一种智能助手，帮助作曲家克服创作瓶颈，获取新鲜灵感。接着，文章介绍了不仅能写歌，还能“倾听”人类同伴并根据其节奏和情绪实时调整演奏的实体机器人。最后文章得出结论：未来音乐产业的主旋律将是人类与机器的协作（collaboration），而非替代，因为音乐背后的人类灵魂永远是不可替代的。 【答案速览】 28. A 29. C 30. B 31. C 【逐题解析】 28. 【答案】A 【考点考向】 细节理解题 【精准定位】 根据题干关键词 concern 和 first paragraph，定位至第一段最后一句：“Will AI and androids soon top the music charts, and could they even replace human musicians entirely？” 【深度解析】 原文首段提到，不仅AI系统能创造原创旋律，科学家还在研发能现场即兴演奏的机器人。这引发了一个紧迫的问题（pressing question）：AI和机器人会很快占据音乐排行榜榜首，甚至“完全取代人类音乐家”吗？选项 A “AI might completely take the place of human musicians（AI可能会完全取代人类音乐家）” 完美对应了原文对于取代人类的这种担忧，故选 A。 【干扰项排雷】 B选项（传统乐队可能会拒绝使用现代技术）在原文中未提及。 C选项（人类音乐家可能会对现场演出失去兴趣）属于主观臆断，无中生有。 D选项（机器创作的音乐对公众来说可能太贵了）在首段未提及成本问题。 29. 【答案】C 【考点考向】 细节理解题 【精准定位】 根据题干关键词 Professor Nick Bryan-Kinns 和 specifically help，定位至第二段。 【深度解析】 原文中 Nick Bryan-Kinns 教授明确指出，AI可以作为智能助手，帮助作曲家“在缺乏灵感时快速获得新鲜的音乐想法（quickly get fresh musical ideas when they run out of inspiration）”，并“克服创作瓶颈（overcome creative blocks）”。选项 C “By providing them with fresh inspiration and overcoming creative blocks（通过为他们提供新鲜的灵感并克服创作瓶颈）” 是对原文的高度同义替换与概括，故选 C。 【干扰项排雷】 A选项（通过在舞台上完美无瑕地演奏乐器）不是教授提到的帮助“人类创作者”的具体方式。 B选项（在他们太累而无法表演时替换他们）与教授“AI不是为了取代人类（Rather than replacing human artists）”的核心观点相悖。 D选项（通过将他们的原创旋律推向音乐排行榜榜首）曲解了原文首段的设问，并非教授提到的实际帮助。 30. 【答案】B 【考点考向】 细节理解题 【精准定位】 根据题干关键词 newly developed androids 和 live performance，定位至第三段第二、三句。 【深度解析】 原文描述这种新研发的实体机器人不仅限于演奏预先编程的音符，它们能够“‘倾听’人类乐队成员的声音并进行现场即兴演奏（can actually "listen" to human bandmates and improvise live music）”。它们利用先进的传感器捕捉人类表演者的节奏和情绪，“从而允许它们实时调整自己的演奏（allowing them to adjust their own playing in real-time）”。选项 B “They can interact with human performers and adjust their playing（它们能与人类表演者互动并调整自己的演奏）” 精准概括了这一特点，故选 B。 【干扰项排雷】 A选项（它们能比任何人类音乐家演奏得更大声）原文未提及音量大小。 C选项（它们能修理人类乐队成员坏掉的乐器）无中生有。 D选项（它们只演奏研究人员编程的准确音符）与原文“don't just play pre-programmed notes”的内容完全相反。 31. 【答案】C 【考点考向】 推理判断题（文章主旨/前景推断） 【精准定位】 根据题干要求推断 future of the music industry，直接定位至最后一段前两句：“Ultimately, the future of the music industry won't be a battle between humans and machines. Instead, it will be about collaboration.” 【深度解析】 结论段明确指出，音乐产业的未来不会是人机大战，而是关于“合作（collaboration）”。虽然AI会塑造未来的音乐，但音乐背后的人类灵魂是不可替代的。选项 C “It will rely heavily on the cooperation between humans and machines（它将严重依赖人类与机器之间的合作）” 中的 cooperation 是对原文 collaboration 的完美同义替换，是对音乐产业未来的正确推断，故选 C。 【干扰项排雷】 A选项（它将完全被AI软件和高级机器人控制）与原文强调的“人类灵魂不可替代”直接矛盾。 B选项（它将见证机器创作音乐的受欢迎程度下降）与首段提到的“机器创作的音乐即将人气飙升（is about to skyrocket in popularity）”相悖。 D选项（它将禁止使用AI工具以保护人类创造力）违背了文章提倡AI作为创新工具来赋能人类（empower human creators）的整体积极基调。 D Recently, a new Chinese AI model called GLM 5.2 has narrowed the gap with American rivals. It promises high capabilities at a fraction of the cost of Western models and has released its core parameters publicly. This has caught global attention, especially as American tech companies are currently grappling with soaring AI costs and strict access restrictions. But is the gap between Chinese and American AI really as small as it seems? While GLM 5.2 ranks highly on public benchmark tests, experts warn of an illusion. In public tests, the gap seems to be just a few months. However, open-source models tend to perform unusually well on these public tests because the exam questions are already published, leading to a phenomenon where labs might unwittingly "teach to the test." On private benchmarks where evaluation questions are kept secret, America’s lead nearly doubles. This indicates that while Chinese AI is improving rapidly, the true capability gap might be larger than public scores suggest. Furthermore, the assumption that Chinese AI is much cheaper can be misleading. While models like GLM 5.2 charge a much lower price per million tokens (bits of text processed by a model), they are generally less efficient. Studies show that Chinese models often use significantly more tokens to think through their answers. Therefore, when comparing the total cost to complete a specific complex task, the Chinese system might end up costing more than its American competitors due to the sheer volume of tokens consumed. Ultimately, as the AI race speeds up, the focus is also shifting to reliability and regulatory risks. While the open-source nature of Chinese models offers a temporary alternative to Western models, Chinese labs face their own challenges, such as computing power shortages. The fact that Chinese models have not yet triggered severe domestic regulatory intervention may be the clearest evidence that, for now, they remain slightly behind the uncontrollable power of their American rivals. 32. What does the comparison between public and private benchmarks reveal? A. Chinese AI models perform better because they process more tokens. B. The actual capability gap between American and Chinese AI is wider than it appears. C. American companies are secretly collecting data from Chinese developers. D. Public tests are no longer used by researchers to evaluate AI intelligence. 33. Why might the Chinese AI end up costing more when completing a specific task? A. It frequently suffers from service interruptions and delays. B. It requires users to purchase highly expensive local hardware. C. It charges a much higher price per token than American AI. D. It processes information less efficiently, consuming far more tokens. 34. What can be inferred about Chinese AI from the last paragraph? A. It is completely immune to any future government regulations. B. It has already alarmed the government due to its uncontrollable power. C. Its current capabilities haven't triggered strict domestic regulatory actions yet. D. It will soon overtake American AI in terms of hardware reliability. 35. What is the author's main purpose in writing this text? A. To promote the global application of open-source AI models. B. To warn American companies about the soaring costs of AI technology. C. To celebrate the remarkable technological breakthroughs of GLM 5.2. D. To offer a critical evaluation of the true competitiveness of Chinese AI. 主题语境：科学与技术 —— 中美人工智能发展对比与客观评估 语篇类型：科技评论 / 议论文 文章大意：本文主要围绕一款名为 GLM 5.2 的新型中国AI大模型展开，探讨了其与美国顶尖AI模型之间的真实差距。尽管该模型在公开基准测试中表现优异、宣称成本极低且核心参数开源，但作者通过三个维度的冷静分析（公开与私有测试集的表现差异、实际Token消耗导致的真实成本，以及国内监管介入程度），指出中国AI与美国AI之间的实际能力差距可能比表面上看起来的更大。文章旨在为读者提供一个客观、批判性的科技发展评估视角。 【答案速览】 32. B 33. D 34. C 35. D 【逐题解析】 32. 【答案】B 【考点考向】 推理判断题（对比细节提取） 【精准定位】 根据题干关键词 public and private benchmarks，定位至第二段最后两句：“On private benchmarks where evaluation questions are kept secret, America’s lead nearly doubles. This indicates that while Chinese AI is improving rapidly, the true capability gap might be larger than public scores suggest.” 【深度解析】 原文指出，在公开测试中差距看似只有几个月，但在考题保密的私有测试（private benchmarks）中，美国的领先优势几乎翻倍。这明确表明，中美AI之间的实际能力差距比公开分数显示的更大。选项 B “The actual capability gap between American and Chinese AI is wider than it appears（美国和中国AI之间的实际能力差距比看起来更宽）” 完美对应了原文的结论，故选 B。 【干扰项排雷】 A选项（中国AI表现更好是因为处理了更多token）：处理更多token是第三段提到的成本增加的原因，与基准测试表现无关。 C选项（美国公司暗中收集中国开发者的资料）和 D选项（研究人员不再使用公开测试）：这两种说法在文中均属于无中生有。 33. 【答案】D 【考点考向】 细节理解题（因果细节溯源） 【精准定位】 根据题干关键词 costing more 和 specific task，定位至第三段最后两句：“Studies show that Chinese models often use significantly more tokens to think through their answers. Therefore, when comparing the total cost to complete a specific complex task, the Chinese system might end up costing more... due to the sheer volume of tokens consumed.” 【深度解析】 文章分析了为何说中国AI更便宜可能是一种误导。虽然它每百万token（文本处理单位）的标价更低，但其效率较低，往往需要消耗大得多的token数量来思考答案。因此，由于消耗了庞大的token总量，完成特定复杂任务的总成本可能反而更高。选项 D “It processes information less efficiently, consuming far more tokens（它处理信息效率较低，消耗了多得多的token）” 精准命中了涨价的根本原因，故选 D。 【干扰项排雷】 A选项（频繁遭受服务中断和延迟）和 B选项（需要用户购买极其昂贵的本地硬件）：文中均未提及这些因素。 C选项（每个token的收费比美国AI高得多）：与原文“charge a much lower price per million tokens（每百万token收费低得多）”的事实完全相反。 34. 【答案】C 【考点考向】 推理判断题 【精准定位】 根据题干关键词 last paragraph，定位至最后一段最后一句：“The fact that Chinese models have not yet triggered severe domestic regulatory intervention may be the clearest evidence that, for now, they remain slightly behind...” 【深度解析】 最后一段提到，中国模型目前尚未引发国内严厉的监管干预（not yet triggered severe domestic regulatory intervention），这反而成了它们目前在能力上仍略逊于美国对手的最清晰证据。由此可以推断出，它们目前的能力水平尚未达到需要政府采取严厉限制措施的程度。选项 C “Its current capabilities haven't triggered strict domestic regulatory actions yet（它目前的能力尚未引发严格的国内监管行动）” 是对原文原句的准确同义转述，故选 C。 【干扰项排雷】 A选项（完全免受未来政府监管影响）：用词“completely immune（完全免疫）”过于绝对，不符合客观事实。 B选项（已经因其不可控的力量惊动了政府）：与原文“have not yet triggered（尚未引发）”相悖。 D选项（硬件可靠性方面即将超越美国）：文中提到了中国实验室面临“计算能力短缺（computing power shortages）”，并未说即将超越。 35. 【答案】D 【考点考向】 主旨大意题（写作意图判断） 【精准定位】 统观全文，重点关注第一段的设问以及后三段的论证方向。 【深度解析】 文章开篇抛出核心疑问：“中国和美国AI的差距真的如看起来那么小吗？（is the gap... really as small as it seems?）”，随后从公开测试成绩存在“应试”水分、低效导致实际使用成本变高，以及未引发严厉监管这三个方面进行了冷静、理性的剖析。显然，作者的核心目的在于打破表面上的繁荣幻象，对中国AI的真实竞争力进行理性的审视。选项 D “To offer a critical evaluation of the true competitiveness of Chinese AI（为了对中国AI的真实竞争力提供批判性评估）” 最精准地概括了作者的写作意图，故选 D。 【干扰项排雷】 A选项（推广开源AI模型全球应用）：文章重点是对比评估，而非商业推广。 B选项（警告美国公司关于AI成本飙升的问题）：虽然第一段提到了美国公司面临成本飙升，但这只是背景引入，而非文章写作目的。 C选项（庆祝GLM 5.2的技术突破）：通读全文可知，文章的基调是“批判性审视与降温”，而非庆祝其技术突破。 学科网（北京）股份有限公司 $ 最新权威外刊改编阅读理解C、D篇（二）【原卷版】 阅读下列短文，从每题所给的 A、B、C、D 四个选项中选出最佳选项。 C More than 80 percent of the world's oceans are currently unmapped. However, a $7 million prize pool named the Ocean Discovery XPrize hopes to change that. The goal is to develop a comprehensive atlas (地图集) of the deep sea by 2030 using uncrewed deep-sea vehicles. The grand prize required entrants to develop an autonomous vessel capable of mapping at least 250 square kilometers of the sea floor within 24 hours, up to a depth of 4 kilometers below the surface. The maps must be highly accurate, with data points taken very closely. The $4 million first prize was awarded to a team whose uncrewed vessel, the SeaKIT, used sonar (声呐) to determine water depth and a cloud-based system that produced an ocean map within 48 hours. A comprehensive map of the world's oceans may uncover new marine species, discover unknown materials, and even find old shipwrecks. "The deep sea is the world's largest museum, and we don't have access to it right now," says Jyotika Virmani, XPrize's Ocean Discovery director. However, exploring the ocean floor is made notoriously difficult by extreme conditions such as complete darkness, freezing temperatures, and high pressure. Beyond creating maps, the technology developed by these teams has other promising applications. For instance, it could be used to track fish populations or monitor invasive species. Furthermore, it could play a vital role in search and rescue missions, such as locating planes that crash into the ocean. With the rapid advancement of deep-sea robots, the hidden secrets of our oceans will eventually be brought to light. 28. What was the specific requirement for the grand prize of the Ocean Discovery XPrize? A. To discover new marine species and ancient shipwrecks. B. To map a specific area of the sea floor within a time limit. C. To track chemical signals underwater back to their source. D. To build a cloud-based system for global weather forecasting. 29. What does the underlined word "autonomous" in paragraph 2 most probably mean? A. Independent. B. Expensive. C. Environmentally-friendly. D. Energy-saving. 30. Why does Jyotika Virmani compare the deep sea to "the world's largest museum"? A. Because it carefully displays the history of human space exploration. B. Because it is well protected by governments from public access. C. Because it holds numerous undiscovered treasures and species. D. Because it requires highly expensive tickets for visitors to enter. 31. What is the text mainly about? A. The dangers of exploring extreme deep-sea environments. B. The historical discoveries of ancient underwater shipwrecks. C. How high school students won a million-dollar science prize. D. How uncrewed deep-sea robots help map the world's oceans. D Brands promoting their products online are quietly deploying AI-generated influencers on social media, an investigation has found, prompting calls for greater transparency. These companies are increasingly turning to AI-generated content that purports (声称) to show genuine customer experiences, while giving no obvious indication that the people featured are not real. The lack of specific rules makes this practice highly controversial. While the EU will soon require AI-manipulated content to be clearly labelled, many other regions currently lack such regulations. A recent investigation found that a worrying 70% of people are unable to correctly identify fake videos. "It is concerning that consumers are not able to trust the content they are seeing online," says Lisa Barber, a technology editor. "Consumers could be frequently misled by AI-generated content and become targets for scammers." Why are brands eagerly embracing this digital deception? The answer lies in economics and control. Brands want high-end marketing without paying tens of thousands of dollars for a traditional photoshoot. AI offers a smarter, more scalable way to create user-generated content. Furthermore, human influencers come with risks like bad press, personal opinions, and high hourly rates. To maintain the illusion of authenticity, creators making AI influencer content are even asked to sign non-disclosure agreements (NDAs) so they cannot reveal the AI origins of the campaigns. Ultimately, while some advertising regulators state that using AI isn't inherently problematic unless the ad itself is misleading, the ethical boundaries remain blurry. As artificial intelligence continues to blur the line between reality and digital fabrication, the core issue of consumer trust is put to the severe test. Companies must prioritize transparency; otherwise, the very authenticity they seek to fake will completely destroy their long-term relationship with consumers. 32. Why do companies favor AI-generated influencers over human ones, according to the text? A. They are more creative in designing brand-new products. B. They strictly follow the advertising rules of the EU. C. They are more cost-effective and easier for brands to control. D. They can interact with target audiences more effectively. 33. What can be inferred from the fact that creators are asked to sign NDAs? A. The AI technology used by the brands is highly confidential and patented. B. Brands intend to conceal the truth to maintain the illusion of authenticity. C. Creators are paid extremely high salaries for their digital artworks. D. Consumers strongly demand to know the personal background of the creators. 34. What is the current regulatory situation regarding AI influencers in most regions outside the EU? A. Brands are explicitly prohibited from using them in advertising. B. Consumers are legally required to report fake promotional videos. C. Advertisements must be strictly labelled as AI-generated. D. There is a lack of specific disclosure rules targeting AI content. 35. What is the author's main purpose in writing the text? A. To introduce a cost-saving trend in social media marketing. B. To highlight the ethical concerns of using undisclosed AI influencers. C. To criticize the poor quality of current user-generated content. D. To encourage consumers to become digital creators on social media. 学科网（北京）股份有限公司 $ 阅读理解C、D篇专项训练 答题卡 姓名：__________________ 班级：__________________ 考号：__________________ 考生须知： 1. 本专项训练仅包含客观选择题（第28-35题）。 2. 请务必使用 2B铅笔 将正确选项的方框涂黑或在方框内打勾。 3. 保持答题卡卷面整洁。 阅读理解填涂区（每题2.5分，共20分） 【C 篇】 28. [ A ] [ B ] [ C ] [ D ] 29. [ A ] [ B ] [ C ] [ D ] 30. [ A ] [ B ] [ C ] [ D ] 31. [ A ] [ B ] [ C ] [ D ] 【D 篇】 32. [ A ] [ B ] [ C ] [ D ] 33. [ A ] [ B ] [ C ] [ D ] 34. [ A ] [ B ] [ C ] [ D ] 35. [ A ] [ B ] [ C ] [ D ] 学科网（北京）股份有限公司 $