热点原创试题01 科学探索中的AI新应用（设计摧毁癌细胞的蛋白质、工作场所偏见识别等）2026年高考英语阅读理解突破策略及押题

热点01.科学探索中的AI新应用 （设计摧毁癌细胞的蛋白质、工作场所偏见识别等） 语篇 内容简介 Passage 1 本文介绍了一场名为“Agents4Science”的会议，该会议旨在测试AI作为科学合作者的能力。 Passage 2 本文报道了一项研究，发现鲸鱼的“咔嗒”声中可能存在“咯咯”声，并推测这可能像人类语言的元音一样传递意义。 Passage 3 本文主要介绍了一项利用人工智能（AI）设计蛋白质来帮助T细胞识别并摧毁癌细胞的新技术。 Passage 4 本文介绍了一种名为Aeneas的人工智能系统，它通过分析拉丁铭文来帮助历史研究。 Passage 5 本文讨论了一项研究，该研究揭示了AI系统如何从有偏见的在线数据中学习到针对女性的工作场所偏见。 Passage 6 本文介绍了一款名为OpenFold3的新型人工智能模型。该模型作为AlphaFold3的开源版本，能够预测蛋白质与其他分子的相互作用，对药物研发至关重要。 Passage 1 A recent conference called Agents4Science was held to explore a new role for artificial intelligence: working as a co-scientist. This virtual meeting was an experiment to see how well AI agents could perform scientific tasks. The research presented covered a wide range of fields, including economics, biology, and engineering, showing the broad interest in this new approach. In this unique setting, AI agents took the lead in many stages of the research process. They were responsible for formulating hypotheses, analyzing data, and even providing initial peer reviews. After the AI's work, human scientists stepped in to review the top submissions. Out of many papers submitted, only a small number were accepted. A key requirement for every accepted paper was to detail how humans and AI collaborated at each step of the research and writing. The results showed both the promise and the problems of using AI in science. On the positive side, AI proved to be very effective at speeding up complex calculations. For example, one economist used AI to study large amounts of data on car-towing policies. However, she also discovered a major drawback: the AI made factual mistakes, such as citing the wrong date for a policy change. This highlights that human oversight is still essential to ensure accuracy. Some experts were not very impressed with the quality of the AI-driven research. One reviewer noted that while the papers she saw were technically correct, they were neither interesting nor important. She worried that the AI's technical skill could hide a lack of good scientific judgment. This suggests that today’s AI may struggle to ask truly meaningful scientific questions on its own. Despite the criticisms, there were positive signs about AI's future in science. In one successful case, a machine learning engineer asked an AI to suggest ideas for a research paper. One of the AI's suggestions was so good that it was selected as one of the top papers at the conference. This example shows that AI might have the ability to come up with truly novel ideas, acting as a creative partner in the scientific process. 1. What is the main purpose of the passage? A. To describe a new conference led entirely by AI. B. To discuss the potential and challenges of AI as a scientific partner. C. To prove that AI is more intelligent than human scientists. D. To encourage all scientific journals to accept AI coauthors. 2. What was a major weakness of the AI-assisted papers according to Risa Wechsler? A. They were too difficult for humans to understand. B. They often contained mathematical errors. C. They lacked originality and significance. D. They were based on incorrect data sources. 3. What can be inferred from Min Min Fong’s experience with AI? A. AI is not yet reliable enough to work independently. B. Economics is the most suitable field for AI application. C. Human scientists are no longer necessary for data analysis. D. AI can perfectly replace human researchers in the future. 4. What positive role did AI play in Silvia Terragni’s project? A. It conducted all the data analysis for her. B. It helped her win the conference competition. C. It generated a winning idea for her research. D. It wrote the final version of her paper. 答案与解析： 文章大意：本文介绍了一场名为“Agents4Science”的会议，该会议旨在测试AI作为科学合作者的能力。文章通过具体案例，展示了AI在加速计算和提出新想法等方面的潜力，同时也指出了AI在事实核查和提出重要科学问题等方面的缺陷。文章最终得出结论：尽管AI是一个有前景的工具，但核心的科研工作仍需人类的判断和监督。 1. B 解析：主旨大意题。 文章开篇介绍了会议的目的是探索AI作为“合作科学家”的角色，接着通过正反两方面的例子，既讨论了AI在加速计算、提出创意方面的潜力，也指出了其在事实准确性、研究重要性方面的挑战。全文围绕AI作为科学伙伴的利弊展开，因此B选项“讨论AI作为科学伙伴的潜力和挑战”最能概括全文主旨。A选项“描述一个完全由AI主导的新会议”与原文不符，会议是人与AI协作。C选项“证明AI比人类科学家更聪明”过于绝对，且与文中提到的AI缺陷相悖。D选项“鼓励所有科学期刊接受AI合作者”并非文章主旨，文章只是客观呈现了现状和探索。 2. C 解析：细节理解题。 根据文章第四段，“One reviewer noted that while the papers she saw were technically correct, they were neither interesting nor important.”（一位审稿人指出，她看到的论文虽然技术上正确，但既无趣也不重要。）C选项“缺乏原创性和重要性”是对这句话的同义转述，其中“originality”对应“interesting”，“significance”对应“important”。A选项“对人类来说太难理解”文中未提及。B选项“经常包含数学错误”是另一位科学家Fong遇到的问题，但Wechsler的评价是“technically correct”。D选项“基于错误的数据源”文中未提及，文中说的是AI引用了错误的日期，属于事实错误，而非数据源错误。 3. A 解析：推理判断题。 根据文章第三段，Fong的经历展示了AI的两面性：一方面“very effective at speeding up complex calculations”（在加速复杂计算方面非常有效），另一方面“made factual mistakes”（犯了事实性错误），并且她强调“human oversight is still essential”（人类的监督仍然至关重要）。由此可以推断出，AI虽然有用，但还不足以独立可靠地工作，仍需人类监督。A选项“AI还不够可靠，无法独立工作”是基于此信息的合理推断。B选项“经济学是AI应用最合适的领域”以偏概全，文章提到多个领域。C选项“人类科学家对于数据分析已不再必要”与原文“human oversight is still essential”的观点完全相反。D选项“AI未来可以完美取代人类研究者”过于武断，与全文的谨慎态度不符。 4. C 解析：细节理解题。 根据文章最后一段，“a machine learning engineer asked an AI to suggest ideas for a research paper. One of the AI’s suggestions was so good that it was selected as one of the top papers at the conference.”（一位机器学习工程师让AI为一篇研究论文提建议。AI的一个建议非常好，被选为会议的最佳论文之一。）这明确说明AI在项目中扮演的积极角色是“为她的研究提供了一个获奖的想法”。C选项“为她的研究产生了一个获奖的想法”是对此信息的准确概括。A选项“为她进行了所有的数据分析”文中未提及。B选项“帮助她赢得了会议竞赛”表述不准确，是AI提出的想法帮助她获奖，而不是AI直接帮助她。D选项“为她写了论文的最终版本”文中未提及。 Passage 2 For a long time, scientists have studied how sperm whales talk to each other using patterns of clicks, known as codas. A recent research project has used artificial intelligence (AI) to listen to these sounds more closely. The team, led by biologist Shane Gero, discovered something surprising. Within the familiar click patterns, they sometimes found what they describe as "clacks." This finding has led to an exciting discussion: could these small changes in sound, like the difference between a click and a clack, carry different meanings for the whales, similar to how humans use different vowels in words? However, this new idea is facing strong disagreement from other experts. Marine biologist Luke Rendell, for example, questions the discovery. He suggests that the "clacks" might not be real at all. Instead, he believes they could be recording errors—unwanted sounds created by the equipment. He points out that there is no proof that the whales themselves are reacting to these sound differences as if they were messages. Without the whales' response, he argues, it's too early to say the sounds are a form of communication. Another expert, Denise Herzing, also has concerns. She worries that comparing whale sounds to human "vowels" is misleading. She fears people might mistakenly believe whales have a human-like language. Herzing recalls that in the past, unproven claims about animal communication harmed scientific research for many years. She believes it's important to be careful and not jump to conclusions without solid evidence. Despite the debate, the research team stands by its work. They mention that they have found the same "clack" pattern in data collected by other scientists using different recording tools, which they believe helps rule out equipment errors. While the argument continues, most agree that using AI to study animal communication is a new and valuable method. This discovery, whether it's a true form of meaning or a technical mistake, has opened a new door for understanding the complex world of whales. 1. What is the main point of disagreement among the scientists? A. The way sperm whales swim in the ocean. B. The function of the "clack" sound in whale communication. C. The best method for recording whale sounds. D. The number of clicks in a sperm whale coda. 2. What can be inferred about Denise Herzing's attitude towards the research? A. She is fully supportive of the new findings. B. She believes the research method is outdated. C. She is cautious about overstating the discovery's importance. D. She thinks the research has no value at all. 3. How did the CETI research team respond to the criticism of recording errors? A. They admitted their equipment was faulty. B. They ignored the criticism from other experts. C. They said other studies with different tools found the same pattern. D. They changed their research method completely. 4. What is the author's main purpose in writing the passage? A. To prove that sperm whales have a complex language. B. To introduce a new AI technology for studying animals. C. To report on a scientific discovery and the debate it caused. D. To compare the communication styles of different whale species. 答案与解析： 文章大意：本文报道了一项利用人工智能研究抹香鲸叫声的新发现。该研究发现鲸鱼的“咔嗒”声中可能存在“咯咯”声，并推测这可能像人类语言的元音一样传递意义。然而，这一发现引发了其他科学家的质疑，他们认为这可能是录音错误或过度解读。文章最后指出，尽管存在争议，但该研究方法为探索动物交流开辟了新途径。 1. B 解析： 本题为主旨大意题中的细节辨析题，考查对文章核心争论点的理解。文章第二段和第三段集中阐述了争议的焦点。Luke Rendell质疑“clack”的真实性，认为其可能是录音错误，并且没有证据表明鲸鱼对此有反应（即不具备交流功能）。Denise Herzing则担心将其比作人类元音会产生误导。这些争论的核心都围绕着“clack”声在鲸鱼交流中到底扮演什么角色，即其“功能”。选项A（游泳方式）和D（咔嗒声数量）与争论无关。选项C（录音方法）虽然被提及，但只是Rendell质疑的论据，而非争论的核心。争论的核心是这个“发现”本身的意义，即B选项。 2. C 解析： 本题为推理判断题。根据第三段，Denise Herzing担心使用“vowel”这个词会“misleading”（误导），让人们误以为鲸鱼拥有类人语言。她还提到过去不成熟的论断曾“harmed scientific research”（伤害了科学研究）。这表明她并不反对研究本身，而是反对在没有确凿证据前就夸大其意义。因此，她的态度是“谨慎的”，担心“过分强调”该发现的重要性。选项A（完全支持）与原文矛盾。选项B（方法过时）和D（毫无价值）在文中没有依据，她认为研究本身是值得探索的。 3. C 解析： 本题为细节理解题。题干询问CETI团队如何回应“录音错误”的批评。文章第四段明确提到：“They mention that they have found the same 'clack' pattern in data collected by other scientists using different recording tools, which they believe helps rule out equipment errors.”（他们提到，在其他科学家用不同录音工具收集的数据中也发现了相同的“咯咯”模式，他们认为这有助于排除设备错误的可能。）这与选项C的表述完全一致。选项A、B、D均与原文信息不符。 4. C 解析： 本题为主旨大意题，考查对作者写作意图的把握。文章首先介绍了一个新发现，然后用大量篇幅呈现了围绕该发现产生的科学辩论，最后给出了一个开放性的结论。整个文章的结构是“提出发现 -> 展示争议 -> 总结展望”。因此，作者的主要目的是“报道一项科学发现及其引发的争议”。选项A（证明鲸鱼有复杂语言）过于绝对，文章只是说“可能”，且争议很大。选项B（介绍AI技术）只是实现发现的手段，不是文章主旨。选项D（比较不同鲸鱼）文章并未涉及。选项C最准确地概括了文章的核心内容。 Passage 3 Scientists are exploring a new way to fight cancer using artificial intelligence (AI). They are using AI to design special proteins that can boost the power of our body's own defense cells, known as T cells, to find and destroy cancer cells. This new method acts like a molecular navigation system, guiding the immune cells directly to their target, much like a GPS guides a traveler to a new address. The challenge in cancer treatment is that T cells, while powerful, sometimes cannot recognize cancer cells effectively. To solve this, researchers turned to AI for help. They genetically engineered T cells to carry tiny, custom-designed proteins on their surface. These man-made proteins serve as a GPS, allowing the T cells to lock onto cancer cells with high precision. This approach is a form of immunotherapy（免疫疗法）, which uses the body's immune system to fight disease, and it is seen as an exciting early test to show the idea can work. The design process is remarkably fast and efficient. The team used a set of AI tools to create these proteins. First, an AI model proposed protein shapes that would perfectly fit the cancer target. Then, another AI model suggested the basic building blocks needed to form those shapes. Finally, a third AI model checked the designs. From tens of thousands of possibilities, the team selected the most promising ones to test in the lab. This entire process can take just a day or two, a huge improvement over older methods that could take months and often produced few useful results. In laboratory experiments, T cells equipped with the AI-designed protein were successful. They were able to quickly kill melanoma cells, a type of skin cancer, and stop the cancer from growing. While this is a significant step forward, experts caution that the work is still in its early stages. Before this treatment can be tried in humans, it will require many more tests in animals and further lab studies, a process that could take several years. Nevertheless, this breakthrough shows the great potential of AI in creating a whole new class of medicines for various diseases in the future. 1. What is the primary role of the AI-designed proteins? A. To directly kill cancer cells. B. To act as a guide for immune cells. C. To replace the body's natural T cells. D. To speed up the growth of cancer cells. 2. What can be inferred about the traditional method of developing cancer treatments? A. It was more successful than the new AI method. B. It was a faster and simpler process. C. It often required a lot of time and effort. D. It was mainly focused on animal testing. 3. What is the author's main purpose in writing this passage? A. To argue against the use of AI in medicine. B. To introduce a new AI-based approach to fighting cancer. C. To compare different types of cancer treatments. D. To explain the history of immunotherapy. 4. What does the passage suggest about the future of this AI technology? A. It is ready for immediate use in hospitals. B. Its application may extend beyond cancer. C. It will be replaced by other technologies soon. D. It has failed to show any positive results. 答案与解析： 文章大意：本文主要介绍了一项利用人工智能（AI）设计蛋白质来帮助T细胞识别并摧毁癌细胞的新技术。文章阐述了该技术的工作原理、AI设计过程的快速高效，以及在实验室中取得的初步成功。同时，文章也指出该研究仍处于早期阶段，需要更多测试，但展示了AI在开发新药物方面的巨大潜力。 1. B 解析：细节理解题。 题目询问AI设计蛋白质的主要作用是什么。根据文章第二段，“These man-made proteins serve as a GPS, allowing the T cells to lock onto cancer cells with high precision.”（这些人造蛋白质充当GPS，让T细胞能够高精度地锁定癌细胞。）以及第一段提到的“guiding the immune cells directly to their target”（直接引导免疫细胞到达目标），可知这些蛋白质的主要作用是作为导航系统，引导T细胞。选项B（作为免疫细胞的向导）准确地概括了这一功能。选项A错误，因为直接杀死癌细胞的是T细胞，而非蛋白质本身。选项C错误，蛋白质是辅助T细胞，而不是取代它们。选项D与文章内容完全相反。 2. C 解析：推理判断题。 题目要求推断关于传统癌症治疗方法的信息。文章第三段将新方法与旧方法进行了对比：“This entire process can take just a day or two, a huge improvement over older methods that could take months and often produced few useful results.”（整个过程只需一两天，这比旧方法是一个巨大进步，旧方法可能需要数月且常常很少产生有用的结果。）由此可以推断，传统方法耗时很长且效率不高。选项C（常常需要大量的时间和精力）是对这一信息的合理推断。选项A错误，文章暗示新方法更成功。选项B错误，文章明确说旧方法更慢、更复杂。选项D在文中没有信息支持，文章说的是新方法未来需要更多动物测试，并未提及旧方法的情况。 3. B 解析：主旨大意题。 题目询问作者写作的主要目的。文章开篇即点明主题“Scientists are exploring a new way to fight cancer using artificial intelligence (AI)”，随后详细介绍了这项技术的原理、过程和前景。全文围绕“利用AI对抗癌症”这一核心展开。因此，选项B（介绍一种基于AI的抗癌新方法）最符合文章主旨。选项A与文章积极、介绍的口吻相反。选项C范围过大，文章主要聚焦于一种新方法，而非广泛比较。选项D过于片面，文章的重点是AI的应用，而非免疫疗法的历史。 4. B 解析：推理判断题。 题目询问文章对这项AI技术未来的暗示。文章最后一段提到：“...this breakthrough shows the great potential of AI in creating a whole new class of medicines for various diseases in the future.”（……这一突破显示了AI在未来为各种疾病创造全新药物种类的巨大潜力。）这句话明确指出，该技术的应用前景不止于癌症，还可以扩展到其他疾病。选项B（其应用可能会扩展到癌症之外）是对此信息的正确推断。选项A错误，文章明确指出还需要数年测试。选项C没有依据，文章对其未来持积极态度。选项D与文章中“significant step forward”（重大进步）的描述相悖。 Passage 4 An advanced artificial intelligence (AI) system called Aeneas is offering a new way to study history, particularly by analyzing ancient Latin writings. This computer program was recently used to examine a famous Roman text, the "Res Gestae Divi Augusti," which describes the deeds of Emperor Augustus. The AI's analysis revealed surprising new details that human experts had previously missed. Aeneas works by searching for similarities in a vast database of Latin inscriptions. It helps historians to interpret, date, and even repair damaged ancient texts. According to its developers, the system is a powerful tool that can handle complex language patterns. For example, when studying the inscription about Augustus, Aeneas found that its language shared subtle connections with Roman legal documents and the special language used to maintain imperial power. This was a fresh insight, showing how the emperor communicated his authority to the public. The AI system also proved useful in solving historical debates. For the Augustus inscription, experts have long disagreed on exactly when it was written. Aeneas suggested two possible time periods: one before the emperor’s death and one shortly after. This result reflects the ongoing discussion among scholars and demonstrates the AI’s ability to model historical uncertainty. It shows that the machine can successfully represent complex academic arguments. For historians, Aeneas is a valuable partner. Studies show that experts who use the system are significantly faster and more accurate in their work. The AI can handle time-consuming tasks, freeing researchers to focus on deeper thinking. Instead of spending countless hours on basic analysis, historians can now spend more time drawing connections across the ancient world. Inscriptions are precious because they offer direct evidence of ancient life, and tools like Aeneas ensure this evidence is understood more fully than ever before. Ultimately, the creators of Aeneas believe it represents more than just a successful project. They see it as a new method for studying the past, one that combines the power of AI with the essential knowledge of human experts. This partnership promises to transform our understanding of history. 1. What is the main purpose of the passage? A. To introduce the life story of the Roman Emperor Augustus. B. To explain how an AI tool is changing historical research. C. To compare the differences between Latin and Greek inscriptions. D. To argue that AI will eventually replace human historians. 2. What can be inferred about the "Res Gestae Divi Augusti" inscription? A. Its exact creation time is still a subject of expert debate. B. It was written entirely by a computer program named Aeneas. C. Its language style was completely unique for its time period. D. It was the first text that Aeneas was ever programmed to study. 3. How does Aeneas directly benefit historians in their work? A. By discovering new, previously unknown Roman inscriptions. B. By making their analytical tasks more efficient and accurate. C. By teaching them the complex Latin language from scratch. D. By providing final answers to all historical uncertainties. 4. What new insight did Aeneas provide about the Augustus inscription? A. It identified the text as a legal document from a Roman court. B. It found the text shared language styles with power-related writings. C. It proved that Augustus wrote the text without any help. D. It showed the text was originally created as a marketplace sign. 答案与解析： 文章大意：本文介绍了一种名为Aeneas的人工智能系统，它通过分析拉丁铭文来帮助历史研究。文章以奥古斯都功绩碑为例，说明了Aeneas不仅发现了文本与法律文件和帝国话语的相似之处，还为历史争议提供了新的视角。最终，文章强调了AI作为一种强大工具，能够提高历史学家的工作效率，并开创了人机结合研究历史的新方法。 1. B 解析：推理判断题。文章开篇即点明Aeneas AI系统为历史研究提供了新方法，并通篇介绍其工作原理、具体应用（分析奥古斯都铭文）、对历史学家的益处以及未来意义。全文的核心是解释这个AI工具如何改变历史研究的模式，因此B选项最符合文章主旨。 2. A 解析：推理判断题。文章第三段提到，对于奥古斯都铭文的创作时间，“experts have long disagreed on exactly when it was written”（专家们长期以来对其确切创作时间存在分歧），而Aeneas的分析结果“reflects the ongoing discussion among scholars”（反映了学者之间持续的讨论）。这可以推断出，关于该铭文的创作时间，专家们至今仍未达成一致，仍存在争议。 3. B 解析：细节理解题。文章第四段明确指出，使用Aeneas系统的专家“are significantly faster and more accurate in their work”（在工作中明显更快、更准确），并且AI能“handle time-consuming tasks”（处理耗时的任务）。这综合起来说明，Aeneas通过使分析任务更高效和更准确来使历史学家受益。 4. B 解析：细节理解题。文章第二段明确指出，Aeneas的一个关键发现是，奥古斯都铭文的语言“shared subtle connections with Roman legal documents and the special language used to maintain imperial power”（与罗马法律文件和用于维护帝国权力的特殊语言有微妙的联系）。B选项中的“power-related writings”（与权力相关的写作）是对原文“the special language used to maintain imperial power”的精炼概括。 Passage 5 A new study shows that AI tools, trained on massive online data, display significant bias against women in the workplace. These AI systems tend to create and favor images of women as younger and less experienced than men. For example, when asked to generate job resumes, an AI model made female candidates appear 1.6 years younger on average. It then ranked these younger female applicants as less qualified than male ones, creating a harmful cycle of bias. This bias does not reflect real-world conditions, where male and female employees are of similar ages. Instead, it is learned from the distorted information available online. The research team analyzed nearly 1.4 million online images and videos. They found that people consistently perceived women in professional pictures as younger than men, especially in high-level jobs like doctors or CEOs. This suggests an online perception that older men, but not older women, are seen as having authority. To further confirm this, the team conducted experiments with people. Participants who searched for and viewed images of female workers, such as mathematicians, later estimated the average age for that profession to be two years younger. The opposite was true when they viewed images of male workers. This experiment proved that the biased information online can directly shape human beliefs about age and gender in jobs. The study then demonstrated that AI models inherit and even worsen these societal biases. When an AI was tasked with creating and ranking thousands of resumes for different jobs, it consistently gave female resumes lower experience levels and lower scores. This AI-driven bias has serious real-world consequences. It may help explain why many companies hire young women but fail to promote them to top positions, creating a "glass ceiling." As society becomes more reliant on AI, such harmful biases could become stronger and more widespread, posing a significant challenge to achieving true equality in the workplace. 1.What is the main idea of the passage? A. AI tools are changing the modern workplace. B. AI systems learn workplace gender bias from online data. C. Women face more difficulties than men in their careers. D. Online pictures are not always true to reality. 2.What can be inferred from the human experiment? A. People trust online images more than text. B. AI systems think in a way similar to humans. C. Young women are better at certain jobs. D. It is difficult to judge a person's age from a picture. 3.Why might AI bias harm women's career development? A. It makes women seem less suitable for top jobs. B. It causes companies to hire fewer women. C. It forces women to retire at an earlier age. D. It shows women are less experienced than men. 4.What is the researchers' attitude towards the future of AI? A. Hopeful. B. Uncertain. C. Worried. D. Supportive. 答案与解析： 文章大意：本文讨论了一项研究，该研究揭示了AI系统如何从有偏见的在线数据中学习到针对女性的工作场所偏见。这些AI工具倾向于将女性描绘得更年轻、经验更少，这可能会强化现实世界中的不平等，并随着AI使用的增加而恶化。 1. B 解析：主旨大意题。文章开篇即点明主旨，指出AI工具从网络数据中学到了针对女性的职场偏见。全文围绕这一核心发现展开，首先说明AI偏见的具体表现，然后分析其来源（网络图片和信息的扭曲），接着通过人类实验和AI实验来证明这一偏见的存在及其影响，最后探讨了其潜在危害。选项B“AI系统从网络数据中学到了职场性别偏见”最准确、最全面地概括了文章的核心内容。 2. B 解析：推理判断题。文章第三段描述了人类实验：人们看到网络图片后，其对职业年龄的认知会受到影响。紧接着第四段开头就说“The study then demonstrated that AI models inherit and worsen these societal biases.”（研究随后证明，AI模型继承并加剧了这些社会偏见）。文章的结构暗示，既然作为人类的人会受图片信息影响而形成偏见，那么同样基于这些海量数据训练的AI系统，其“思维”方式（信息处理模式）与人类有相似之处，因此也学会了同样的偏见。这是一个基于文章逻辑链条的合理推断。 3. A 解析：细节理解题。题目询问AI偏见为何会损害女性的职业发展。文章第四段明确指出，AI给女性简历打更低的分数，并解释了其后果：“This AI-driven bias has serious real-world consequences. It may help explain why many companies hire young women but fail to promote them to top positions, creating a ‘glass ceiling’.”（这种AI驱动的偏见有严重的现实后果。它可能有助于解释为什么许多公司雇佣年轻女性却未能将她们提拔到高层职位，从而造成了“玻璃天花板”）。将女性简历评价为“less qualified”（资历较浅），直接导致她们在晋升竞争中处于不利地位，即“seem less suitable for top jobs”（看起来不太适合顶级工作）。选项A是对此原因的精准概括。 4. C 解析： 推理判断题（作者/研究者态度题）。文章结尾部分明确表达了研究者的担忧。文章提到AI偏见是“harmful”（有害的），并且最后一句指出“as society becomes more reliant on AI, such harmful biases could become stronger and more widespread”（随着社会越来越依赖AI，这种有害的偏见可能会变得更强、更普遍）。这些措辞清晰地表明了研究者对未来AI发展可能带来的负面影响的忧虑。因此，他们的态度是“worried”（担忧的）。 Passage 6 A new AI model named OpenFold3 is making significant progress in predicting how proteins interact with other molecules. This open-source version is a careful reconstruction of Google DeepMind’s AlphaFold3. While AlphaFold3’s use is limited, OpenFold3 is available to everyone, including companies, for commercial purposes like drug development. This breakthrough allows scientists to see how proteins pair with other substances, which is shown in computer images that compare predictions with real structures. Predicting these interactions is extremely important for designing new medicines. As one expert explained, biology functions through the constant interaction between different biomolecules, not through proteins working alone. Understanding these pairings is therefore key to creating drugs that can target specific proteins to treat diseases. Proteins are essential workers in the body, and their function depends heavily on their shape and how they connect with other molecules. The creation of OpenFold3 was driven by a need for openness in science. Unlike its predecessor, AlphaFold2, the code for AlphaFold3 was not initially shared with the research community. This made it difficult for scientists to independently verify the model’s accuracy and reliability. After many researchers called for transparency, a team successfully studied and rebuilt the complex system, leading to the development of OpenFold3. Even with its success, OpenFold3 still faces challenges. Perfectly copying a complex AI model is difficult because some small but important adjustments are not written in the code. Furthermore, the current model creates static images, which do not fully capture the dynamic reality inside cells where proteins are surrounded by water and are constantly in motion. The team behind OpenFold3 hopes to add these natural elements in the future to make its predictions even more accurate. The model is already being put to practical use. Several drug companies have formed a cooperative group to train OpenFold3 on their own private data. This method allows them to build a more powerful tool together without sharing their sensitive information. Each company uses its unique data to improve the model, and the combined knowledge is then used to create a stronger, globally enhanced version, accelerating the discovery of new life-saving medicines. 1. Why is predicting protein-molecule interactions crucial for drug design? A. Because proteins are the hardest working molecules. B. Because biology works through molecular interactions. C. Because AlphaFold3 is a closed and limited model. D. Because companies need to develop new drugs quickly. 2. What can be inferred about OpenFold3's current limitation? A. It cannot be used for commercial drug development. B. Its predictions are less accurate than AlphaFold3’s. C. It does not fully represent the natural state of proteins. D. It requires a huge amount of public data to function. 3. What was the primary motivation for researchers to rebuild AlphaFold3? A. To win a Nobel Prize for their scientific achievement. B. To test the model’s accuracy and understand its workings. C. To create a more profitable AI platform for businesses. D. To prove that AI can memorize protein structures. 4. What is the main purpose of the passage? A. To compare the functions of AlphaFold2 and AlphaFold3. B. To introduce OpenFold3 and its significance in science. C. To argue that AI models should always be open-source. D. To explain the basic process of how proteins fold. 答案与解析： 文章大意：本文介绍了一款名为OpenFold3的新型人工智能模型。该模型作为AlphaFold3的开源版本，能够预测蛋白质与其他分子的相互作用，对药物研发至关重要。文章阐述了OpenFold3的诞生背景、相较于AlphaFold3的开放性优势、当前面临的局限性，以及其在制药领域的实际应用和未来发展方向，突出了其在推动科学进步和医学发展方面的巨大潜力。 1.B 解析：细节理解题。本题询问为什么预测蛋白质-分子相互作用对药物设计至关重要。文章第二段明确指出：“Predicting these interactions is extremely important for designing new medicines. As one expert explained, biology functions through the constant interaction between different biomolecules, not through proteins working alone.” 这说明，其重要性在于生物学本身就是通过分子间的相互作用来运作的。 2.C 解析：推理判断题。题要求推断OpenFold3当前的局限性是什么。文章第四段提到：“Furthermore, the current model creates static images, which do not fully capture the dynamic reality inside cells where proteins are surrounded by water and are constantly in motion.” 这句话暗示了当前模型是静态的，未能完全反映细胞内蛋白质的真实动态环境。 3.B 解析：细节理解题。本题询问研究人员重建AlphaFold3的主要动机是什么。文章第三段指出：“This made it difficult for scientists to independently verify the model’s accuracy and reliability.” 以及“a team successfully studied and rebuilt the complex system”。这表明，重建的主要目的是为了验证模型的准确性、可靠性并理解其工作原理。 4.B 解析：主旨大意题。本题要求判断文章的主要写作目的。文章从开头介绍OpenFold3是什么，到中间解释其重要性、诞生背景和局限性，再到最后讲述其应用，全文围绕OpenFold3展开。 原创精品资源学科网独家享有版权，侵权必究！1 学科网（北京）股份有限公司1 / 1 学科网（北京）股份有限公司 $ 热点01.科学探索中的AI新应用 （设计摧毁癌细胞的蛋白质、工作场所偏见识别等） 语篇 内容简介 Passage 1 本文介绍了一场名为“Agents4Science”的会议，该会议旨在测试AI作为科学合作者的能力。 Passage 2 本文报道了一项研究，发现鲸鱼的“咔嗒”声中可能存在“咯咯”声，并推测这可能像人类语言的元音一样传递意义。 Passage 3 本文主要介绍了一项利用人工智能（AI）设计蛋白质来帮助T细胞识别并摧毁癌细胞的新技术。 Passage 4 本文介绍了一种名为Aeneas的人工智能系统，它通过分析拉丁铭文来帮助历史研究。 Passage 5 本文讨论了一项研究，该研究揭示了AI系统如何从有偏见的在线数据中学习到针对女性的工作场所偏见。 Passage 6 本文介绍了一款名为OpenFold3的新型人工智能模型。该模型作为AlphaFold3的开源版本，能够预测蛋白质与其他分子的相互作用，对药物研发至关重要。 Passage 1 A recent conference called Agents4Science was held to explore a new role for artificial intelligence: working as a co-scientist. This virtual meeting was an experiment to see how well AI agents could perform scientific tasks. The research presented covered a wide range of fields, including economics, biology, and engineering, showing the broad interest in this new approach. In this unique setting, AI agents took the lead in many stages of the research process. They were responsible for formulating hypotheses, analyzing data, and even providing initial peer reviews. After the AI's work, human scientists stepped in to review the top submissions. Out of many papers submitted, only a small number were accepted. A key requirement for every accepted paper was to detail how humans and AI collaborated at each step of the research and writing. The results showed both the promise and the problems of using AI in science. On the positive side, AI proved to be very effective at speeding up complex calculations. For example, one economist used AI to study large amounts of data on car-towing policies. However, she also discovered a major drawback: the AI made factual mistakes, such as citing the wrong date for a policy change. This highlights that human oversight is still essential to ensure accuracy. Some experts were not very impressed with the quality of the AI-driven research. One reviewer noted that while the papers she saw were technically correct, they were neither interesting nor important. She worried that the AI's technical skill could hide a lack of good scientific judgment. This suggests that today’s AI may struggle to ask truly meaningful scientific questions on its own. Despite the criticisms, there were positive signs about AI's future in science. In one successful case, a machine learning engineer asked an AI to suggest ideas for a research paper. One of the AI's suggestions was so good that it was selected as one of the top papers at the conference. This example shows that AI might have the ability to come up with truly novel ideas, acting as a creative partner in the scientific process. 1. What is the main purpose of the passage? A. To describe a new conference led entirely by AI. B. To discuss the potential and challenges of AI as a scientific partner. C. To prove that AI is more intelligent than human scientists. D. To encourage all scientific journals to accept AI coauthors. 2. What was a major weakness of the AI-assisted papers according to Risa Wechsler? A. They were too difficult for humans to understand. B. They often contained mathematical errors. C. They lacked originality and significance. D. They were based on incorrect data sources. 3. What can be inferred from Min Min Fong’s experience with AI? A. AI is not yet reliable enough to work independently. B. Economics is the most suitable field for AI application. C. Human scientists are no longer necessary for data analysis. D. AI can perfectly replace human researchers in the future. 4. What positive role did AI play in Silvia Terragni’s project? A. It conducted all the data analysis for her. B. It helped her win the conference competition. C. It generated a winning idea for her research. D. It wrote the final version of her paper. Passage 2 For a long time, scientists have studied how sperm whales talk to each other using patterns of clicks, known as codas. A recent research project has used artificial intelligence (AI) to listen to these sounds more closely. The team, led by biologist Shane Gero, discovered something surprising. Within the familiar click patterns, they sometimes found what they describe as "clacks." This finding has led to an exciting discussion: could these small changes in sound, like the difference between a click and a clack, carry different meanings for the whales, similar to how humans use different vowels in words? However, this new idea is facing strong disagreement from other experts. Marine biologist Luke Rendell, for example, questions the discovery. He suggests that the "clacks" might not be real at all. Instead, he believes they could be recording errors—unwanted sounds created by the equipment. He points out that there is no proof that the whales themselves are reacting to these sound differences as if they were messages. Without the whales' response, he argues, it's too early to say the sounds are a form of communication. Another expert, Denise Herzing, also has concerns. She worries that comparing whale sounds to human "vowels" is misleading. She fears people might mistakenly believe whales have a human-like language. Herzing recalls that in the past, unproven claims about animal communication harmed scientific research for many years. She believes it's important to be careful and not jump to conclusions without solid evidence. Despite the debate, the research team stands by its work. They mention that they have found the same "clack" pattern in data collected by other scientists using different recording tools, which they believe helps rule out equipment errors. While the argument continues, most agree that using AI to study animal communication is a new and valuable method. This discovery, whether it's a true form of meaning or a technical mistake, has opened a new door for understanding the complex world of whales. 1. What is the main point of disagreement among the scientists? A. The way sperm whales swim in the ocean. B. The function of the "clack" sound in whale communication. C. The best method for recording whale sounds. D. The number of clicks in a sperm whale coda. 2. What can be inferred about Denise Herzing's attitude towards the research? A. She is fully supportive of the new findings. B. She believes the research method is outdated. C. She is cautious about overstating the discovery's importance. D. She thinks the research has no value at all. 3. How did the CETI research team respond to the criticism of recording errors? A. They admitted their equipment was faulty. B. They ignored the criticism from other experts. C. They said other studies with different tools found the same pattern. D. They changed their research method completely. 4. What is the author's main purpose in writing the passage? A. To prove that sperm whales have a complex language. B. To introduce a new AI technology for studying animals. C. To report on a scientific discovery and the debate it caused. D. To compare the communication styles of different whale species. Passage 3 Scientists are exploring a new way to fight cancer using artificial intelligence (AI). They are using AI to design special proteins that can boost the power of our body's own defense cells, known as T cells, to find and destroy cancer cells. This new method acts like a molecular navigation system, guiding the immune cells directly to their target, much like a GPS guides a traveler to a new address. The challenge in cancer treatment is that T cells, while powerful, sometimes cannot recognize cancer cells effectively. To solve this, researchers turned to AI for help. They genetically engineered T cells to carry tiny, custom-designed proteins on their surface. These man-made proteins serve as a GPS, allowing the T cells to lock onto cancer cells with high precision. This approach is a form of immunotherapy（免疫疗法）, which uses the body's immune system to fight disease, and it is seen as an exciting early test to show the idea can work. The design process is remarkably fast and efficient. The team used a set of AI tools to create these proteins. First, an AI model proposed protein shapes that would perfectly fit the cancer target. Then, another AI model suggested the basic building blocks needed to form those shapes. Finally, a third AI model checked the designs. From tens of thousands of possibilities, the team selected the most promising ones to test in the lab. This entire process can take just a day or two, a huge improvement over older methods that could take months and often produced few useful results. In laboratory experiments, T cells equipped with the AI-designed protein were successful. They were able to quickly kill melanoma cells, a type of skin cancer, and stop the cancer from growing. While this is a significant step forward, experts caution that the work is still in its early stages. Before this treatment can be tried in humans, it will require many more tests in animals and further lab studies, a process that could take several years. Nevertheless, this breakthrough shows the great potential of AI in creating a whole new class of medicines for various diseases in the future. 1. What is the primary role of the AI-designed proteins? A. To directly kill cancer cells. B. To act as a guide for immune cells. C. To replace the body's natural T cells. D. To speed up the growth of cancer cells. 2. What can be inferred about the traditional method of developing cancer treatments? A. It was more successful than the new AI method. B. It was a faster and simpler process. C. It often required a lot of time and effort. D. It was mainly focused on animal testing. 3. What is the author's main purpose in writing this passage? A. To argue against the use of AI in medicine. B. To introduce a new AI-based approach to fighting cancer. C. To compare different types of cancer treatments. D. To explain the history of immunotherapy. 4. What does the passage suggest about the future of this AI technology? A. It is ready for immediate use in hospitals. B. Its application may extend beyond cancer. C. It will be replaced by other technologies soon. D. It has failed to show any positive results. Passage 4 An advanced artificial intelligence (AI) system called Aeneas is offering a new way to study history, particularly by analyzing ancient Latin writings. This computer program was recently used to examine a famous Roman text, the "Res Gestae Divi Augusti," which describes the deeds of Emperor Augustus. The AI's analysis revealed surprising new details that human experts had previously missed. Aeneas works by searching for similarities in a vast database of Latin inscriptions. It helps historians to interpret, date, and even repair damaged ancient texts. According to its developers, the system is a powerful tool that can handle complex language patterns. For example, when studying the inscription about Augustus, Aeneas found that its language shared subtle connections with Roman legal documents and the special language used to maintain imperial power. This was a fresh insight, showing how the emperor communicated his authority to the public. The AI system also proved useful in solving historical debates. For the Augustus inscription, experts have long disagreed on exactly when it was written. Aeneas suggested two possible time periods: one before the emperor’s death and one shortly after. This result reflects the ongoing discussion among scholars and demonstrates the AI’s ability to model historical uncertainty. It shows that the machine can successfully represent complex academic arguments. For historians, Aeneas is a valuable partner. Studies show that experts who use the system are significantly faster and more accurate in their work. The AI can handle time-consuming tasks, freeing researchers to focus on deeper thinking. Instead of spending countless hours on basic analysis, historians can now spend more time drawing connections across the ancient world. Inscriptions are precious because they offer direct evidence of ancient life, and tools like Aeneas ensure this evidence is understood more fully than ever before. Ultimately, the creators of Aeneas believe it represents more than just a successful project. They see it as a new method for studying the past, one that combines the power of AI with the essential knowledge of human experts. This partnership promises to transform our understanding of history. 1. What is the main purpose of the passage? A. To introduce the life story of the Roman Emperor Augustus. B. To explain how an AI tool is changing historical research. C. To compare the differences between Latin and Greek inscriptions. D. To argue that AI will eventually replace human historians. 2. What can be inferred about the "Res Gestae Divi Augusti" inscription? A. Its exact creation time is still a subject of expert debate. B. It was written entirely by a computer program named Aeneas. C. Its language style was completely unique for its time period. D. It was the first text that Aeneas was ever programmed to study. 3. How does Aeneas directly benefit historians in their work? A. By discovering new, previously unknown Roman inscriptions. B. By making their analytical tasks more efficient and accurate. C. By teaching them the complex Latin language from scratch. D. By providing final answers to all historical uncertainties. 4. What new insight did Aeneas provide about the Augustus inscription? A. It identified the text as a legal document from a Roman court. B. It found the text shared language styles with power-related writings. C. It proved that Augustus wrote the text without any help. D. It showed the text was originally created as a marketplace sign. Passage 5 A new study shows that AI tools, trained on massive online data, display significant bias against women in the workplace. These AI systems tend to create and favor images of women as younger and less experienced than men. For example, when asked to generate job resumes, an AI model made female candidates appear 1.6 years younger on average. It then ranked these younger female applicants as less qualified than male ones, creating a harmful cycle of bias. This bias does not reflect real-world conditions, where male and female employees are of similar ages. Instead, it is learned from the distorted information available online. The research team analyzed nearly 1.4 million online images and videos. They found that people consistently perceived women in professional pictures as younger than men, especially in high-level jobs like doctors or CEOs. This suggests an online perception that older men, but not older women, are seen as having authority. To further confirm this, the team conducted experiments with people. Participants who searched for and viewed images of female workers, such as mathematicians, later estimated the average age for that profession to be two years younger. The opposite was true when they viewed images of male workers. This experiment proved that the biased information online can directly shape human beliefs about age and gender in jobs. The study then demonstrated that AI models inherit and even worsen these societal biases. When an AI was tasked with creating and ranking thousands of resumes for different jobs, it consistently gave female resumes lower experience levels and lower scores. This AI-driven bias has serious real-world consequences. It may help explain why many companies hire young women but fail to promote them to top positions, creating a "glass ceiling." As society becomes more reliant on AI, such harmful biases could become stronger and more widespread, posing a significant challenge to achieving true equality in the workplace. 1.What is the main idea of the passage? A. AI tools are changing the modern workplace. B. AI systems learn workplace gender bias from online data. C. Women face more difficulties than men in their careers. D. Online pictures are not always true to reality. 2.What can be inferred from the human experiment? A. People trust online images more than text. B. AI systems think in a way similar to humans. C. Young women are better at certain jobs. D. It is difficult to judge a person's age from a picture. 3.Why might AI bias harm women's career development? A. It makes women seem less suitable for top jobs. B. It causes companies to hire fewer women. C. It forces women to retire at an earlier age. D. It shows women are less experienced than men. 4.What is the researchers' attitude towards the future of AI? A. Hopeful. B. Uncertain. C. Worried. D. Supportive. Passage 6 A new AI model named OpenFold3 is making significant progress in predicting how proteins interact with other molecules. This open-source version is a careful reconstruction of Google DeepMind’s AlphaFold3. While AlphaFold3’s use is limited, OpenFold3 is available to everyone, including companies, for commercial purposes like drug development. This breakthrough allows scientists to see how proteins pair with other substances, which is shown in computer images that compare predictions with real structures. Predicting these interactions is extremely important for designing new medicines. As one expert explained, biology functions through the constant interaction between different biomolecules, not through proteins working alone. Understanding these pairings is therefore key to creating drugs that can target specific proteins to treat diseases. Proteins are essential workers in the body, and their function depends heavily on their shape and how they connect with other molecules. The creation of OpenFold3 was driven by a need for openness in science. Unlike its predecessor, AlphaFold2, the code for AlphaFold3 was not initially shared with the research community. This made it difficult for scientists to independently verify the model’s accuracy and reliability. After many researchers called for transparency, a team successfully studied and rebuilt the complex system, leading to the development of OpenFold3. Even with its success, OpenFold3 still faces challenges. Perfectly copying a complex AI model is difficult because some small but important adjustments are not written in the code. Furthermore, the current model creates static images, which do not fully capture the dynamic reality inside cells where proteins are surrounded by water and are constantly in motion. The team behind OpenFold3 hopes to add these natural elements in the future to make its predictions even more accurate. The model is already being put to practical use. Several drug companies have formed a cooperative group to train OpenFold3 on their own private data. This method allows them to build a more powerful tool together without sharing their sensitive information. Each company uses its unique data to improve the model, and the combined knowledge is then used to create a stronger, globally enhanced version, accelerating the discovery of new life-saving medicines. 1. Why is predicting protein-molecule interactions crucial for drug design? A. Because proteins are the hardest working molecules. B. Because biology works through molecular interactions. C. Because AlphaFold3 is a closed and limited model. D. Because companies need to develop new drugs quickly. 2. What can be inferred about OpenFold3's current limitation? A. It cannot be used for commercial drug development. B. Its predictions are less accurate than AlphaFold3’s. C. It does not fully represent the natural state of proteins. D. It requires a huge amount of public data to function. 3. What was the primary motivation for researchers to rebuild AlphaFold3? A. To win a Nobel Prize for their scientific achievement. B. To test the model’s accuracy and understand its workings. C. To create a more profitable AI platform for businesses. D. To prove that AI can memorize protein structures. 4. What is the main purpose of the passage? A. To compare the functions of AlphaFold2 and AlphaFold3. B. To introduce OpenFold3 and its significance in science. C. To argue that AI models should always be open-source. D. To explain the basic process of how proteins fold. 原创精品资源学科网独家享有版权，侵权必究！1 学科网（北京）股份有限公司1 / 1 学科网（北京）股份有限公司 $