热点02 Deepseek人工智能AI-2025年高考英语【热点·重点·难点】专练（天津专用）

热点02 DeepSeek人工智能AI 在科技浪潮汹涌澎湃的今天，人工智能AI正以前所未有的速度重塑着我们的世界。在这股浪潮中，DeepSeek如同一颗璀璨的新星，以其卓越的智能解决方案，正引领着未来的方向。DeepSeek不仅仅是一款AI模型，它是智慧与创新的结晶，以其独特的魅力在人工智能领域独树一帜。让我们一同探索DeepSeek的奥秘，见证它如何以其卓越的智能，为我们的生活带来革命性的变化，开启智能科技的新纪元。 （建议用时：60分钟） 人工智能对人类的影响 Passage 1 （2024·天津十二区重点校·一模）假设你是晨光中学高三学生李津。你有一位英国好友Jim，他告诉你他最近经常使用ChatGPT（一款人工智能的语言类软件）来完成他的学习任务，比如写演讲稿和实验报告。他问你是否愿意使用这类人工智能来完成学习任务，你写给他一封E-mail，主要内容包括： 1. 你是否愿意让人工智能替你完成学习任务及其原因。 2. 你在生活中是怎样利用高科技产品的。 3. 祝愿Jim学业进步。 注意：1. 词数100左右：2. 开头和结尾已给出，不计入总词数。 词汇提示： 人工智能：AI (Artificial Intelligence)  高科技产品：high-tech products Dear Jim, ____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ Yours, Li Jin Passage 2 （2023·天津·模拟预测）假设你是振华中学高三学生李明。你与交换生James要共同完成一篇项目学习的论文，他建议用AI完成初稿。请你用英语给他写一封电子邮件，内容包括： 1. 委婉拒绝并说明理由； 2. 你的建议及计划。 注意：1. 词数100左右； 2. 开头和结尾已给出，不计入总词数。 提示词语：项目学习 project-based learning Dear James, ____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ Yours, Li Ming Passage 3 （24-25高三上·天津·阶段练习）A study conducted by Cornell University examined how the use of AI in conversations impacts the way people express themselves and view each other. The researchers have found people have more efficient conversations, use more positive language and perceive each other more positively when using an AI-enabled chat tool. However the group also found that when participants think their partner is using more AI-suggested responses, they consider that partner as less cooperative. “I was surprised to find people tend to evaluate you more negatively simply because they suspect you’re using AI to help compose text, regardless of whether you actually are,” said Jess Hohenstein, the lead researcher. “This illustrates the continuous overall doubt that people seem to have around AI.” For their first experiment, participants were asked to talk about a policy issue and assigned to one of three conditions: both participants can use smart replies; only one participant can use smart replies; or neither participant can use smart replies. Researchers found that using smart replies increased communication efficiency, positive emotional language and positive evaluations by communication partners. On average, smart replies accounted for 14.3% of sent messages. But participants whose partners suspected of responding with smart replies were evaluated more negatively than those who were thought to have typed their own responses, consistent with common assumptions about the negative implications of AI. “While AI might be able to help you write,” Hohenstein said, “it’s altering your language in ways you might not expect, especially by making you sound more positive. This suggests that by, using text-generating AI, you’re giving up some of your-own personal voice.” Malte Jung, an associate professor, said, “What we observe in this study is the impact that AI has on social dynamics and some of the unintended consequences that could result from integrating AI in social contexts. This suggests that whoever controls the algorithm (算法) may have influence on people’s interactions, language and insights into each other.” 6．Which of the following statements does Jess Hohenstein agree with? A．AI always expresses in ways you expect. B．Algorithm will never influence people’s insights. C．Trust can be affected by using AI in conversation. D．You will regain your voice by using AI in conversation. 7．How do the researchers draw their conclusion? A．By analyzing figures. B．By performing a survey. C．By conducting experiments. D．By making comparisons. 8．How will a person feel about suspecting his partner’s using smart replies? A．Nervous. B．Uncomfortable. C．Excited. D．Puzzled. 9．According to the passage, what are the unexpected consequences of using text- generating AI? A．Your language remains unchanged. B．It doesn’t alter speakers’ insights into each other. C．It makes the speaker sound more negative. D．You are losing some of your personal style. 5．What is the passage primarily about? A．Methods of using AI in conversations. B．Efficiency of using AI in conversations. C．Convenience of using AI in conversations. D．Impacts of using AI in conversations. Passage 4 （2024·天津九校·二模）After astonishing breakthroughs in artificial intelligence, many people worry that they will be replaced by AI. Global Google searches for “is my job safe?” have doubled in recent months, as people fear that they will be replaced with large language models (LLMs). In a recent paper Tyna Eloundou of OpenAI and colleagues say that “around 80% of the US workforce could have at least 10% of their work tasks affected by the introduction of LLMs”. Economists, however, tend to enjoy making predictions about automation more than they enjoy testing them. In the early 2010s many of them loudly predicted that robots would kill jobs by the millions, only to fall silent when employment rates across the rich world rose to all-time highs. Few of the doom-mongers (末日论者) have a good explanation for why countries with the highest rates of tech usage around the globe, such as Japan, Singapore and the Republic of Korea, consistently have among the lowest rates of unemployment. Here we introduce our first attempt at tracking AI’s impact on jobs. Using American data on employment by occupation, we single out white-collar workers. White-collar roles are thought to be especially sensitive to generative AI, which is becoming ever better at logical reasoning and creativity. However, there is as yet little evidence of an AI hit to employment. In the spring of 2020 white-collar jobs rose, as many people in service occupations lost their job at the start of the Covid-19 pandemic. And in the past year the white-collar share of employment in professions thought to be at risk from generative AI has risen by 0.5%. Of course, it’s still early days for generative AI. Few firms yet use generative Al tools at large scale, so the impact on jobs could merely be delayed. Another possibility, however, is that these new technologies will end up destroying only a small number of roles. While Al may be efficient at some tasks, it may be less good at others, such as management and working out what others need. AI could even have a positive effect on jobs. If workers using it become more efficient, profits at their company could rise which would then allow bosses to ramp up hiring. A recent survey by Experis, an IT-recruitment firm, points to this possibility. More than half of Britain’s employers expect AI technologies to have a positive impact on the number of their staff over the next two years, it finds. To see how it all shakes out, we will publish updates to this analysis every few months. But for now, a job apocalypse seems a long way off. 1．In terms of artificial intelligence, people worry about ______. A．how to avoid being replaced by machines B．how to keep up with the trend of technology C．how to master new technological skills D．how to earn higher wages by using technology 2．Economists’ predictions about automation show that ______. A．machines prove to deal a blow to employment B．technological breakthroughs bring an economic boom C．economic theories don’t work well in practice D．it is difficult to clarify the impact of machines on jobs 3．According to the text, since the spring of 2020, white-collar jobs ______. A．have become more diverse B．have decreased in number C．have been threatened by service jobs D．have not suffered noticeable losses 4．How may artificial intelligence influence employment? A．It may cause mass unemployment. B．It may create new jobs for people. C．It may enable employers to hire more. D．It may decrease people’s salaries. 5．What could be the best title for the passage? A．Boy Cries Wolf: Generative Al is not yet killing jobs B．Batten Down the Hatches: Generative AI will replace most of human jobs C．Generative AI: The greatest threat to white-collar workers. D．Generative AI: The greatest invention of the 20th century 人工智能工具介绍 Passage 1 （24-25高三上·天津南开·阶段练习）Imagine this. You need an image of a balloon for a work presentation and turn to an Al text-to-image generator, like Midjourney or DALL-E, to create a suitable image. You enter the prompt (提示词) “red balloon against a blue sky” but the generator returns an image of an egg instead. What’s going on? The generator you’re using may have been “poisoned”. What does this mean? Text-to-image generators work by being trained on large datasets that include millions or billions of images. Some of the generators have been trained by — indiscriminately scraping online images, many of which may be under copyright. This has led to many copyright infringement (侵害) cases where artists have accused big tech companies of stealing and profiting from their work. This is also where the idea of “poison” comes in. Researchers who want to empower individual artists have recently created a tool named “Nightshade” to fight back against unauthorised image scraping. The tool works by slightly altering an image’s pixels (像素) in a way that confuses the computer vision system but leaves the image unaltered to a human’s eyes. If an organization then scrapes one of these images to train a future AI model, its data pool becomes “poisoned”. This can result in mistaken learning, which makes the generator return unintended results. As in our earlier example, a balloon might become an egg. The higher the number of “poisoned” images in the training data, the greater the impact. Because of how generative AI works, the damage from “poisoned” images also affects related prompt keywords. For example, if a “poisoned” image of a Picasso work is used in training data, prompt results for masterpieces from other artists can also be affected. Possibly, tools like Nightshade can be abused by some users to intentionally upload “poisoned” images in order to confuse AI generators. But the Nightshade’s developer hopes the tool will make big tech companies more respectful of copyright. It does challenge a common belief among computer scientists that data found online can be used for any purpose they see fit. Human rights activists, for example, have been concerned for some time about the indiscriminate use of machine vision in wider society. This concern is particularly serious concerning facial recognition. There is a clear connection between facial recognition cases and data poisoning, as both relate to larger questions around technological governance. It may be better to see data poisoning as an innovative solution to the denial of some fundamental human rights. 1．The first paragraph is used to ____________. A．report a piece of news B．inform an academic accident C．arouse readers’ interest D．introduce some AI generators 2．The underlined word ‘scraping’ (para. 2) is closest in meaning to ____________. A．facilitating B．collecting C．damaging D．polishing 3．According to the passage, adding poisoned data might ____________. A．increase the accuracy of returned information B．cause users to forget the prompt key words C．interfere with the training of generative AI D．discriminate against great masterpieces 4．What can be inferred from the last two paragraphs? A．Data poisoning is somehow justified to direct attention to human rights. B．Computer scientists have learned to respect the copyright of most artists. C．Nightshade is being abused by human rights activists to recognize faces. D．The issue of technological governance has aroused the lawyers’ interest. 5．Which of the following might be the best title of the passage? A．Data Poisoning: Government Empowering Citizens to Protect Themselves B．Data Poisoning: Addressing Facial Recognition Issues Among Artists C．Data Poisoning: Risks and Rewards of Generative AI Data Training D．Data Poisoning: Restricting Innovation or Empowering Artists Passage 2 （2024·天津河东·二模）After waking up, you may feel frustrated that you cannot recall the dreams you had last night. Artificial intelligence (AI) may be able to help. Previously, there have been AI models that can turn text into images. They can do this by learning from a large amount of data from both tests and images. This time, researchers from Osaka University in Japan have trained an AI system called Stable Diffusion to re-create images based on people’s brain scans, reported Science magazine. The researchers used an online data set provided by the University of Minnesota, US, which consisted of brain scans from four participants as they each viewed a set of 10,000 photos. The scans were recorded by functional magnetic resonance imaging (FMRI,功能性磁共振成像). The AI then learned about the brain activities by analyzing changes in blood flow shown by the FMRI data - when a part of the brain is activated, more blood will flow to it. It then matched the brain activities with the photos. Through this method, AI learned how human brains would react when seeing different photos. Finally, the researchers tested the AI on additional brain scans from the same participants when they viewed photos of a toy bear, airplane, clock and train. If the person looked at an airplane, for example, the AI would use the brain scan data to create an image of a very blurry (模糊的) airplane. Then, it would turn on the previous “text-to-image” model and improve the quality of the image by feeding itself the keyword “airplane”. The final images were “convincing” with about 80 percent of accuracy, according to the researchers. The new study created a novel approach that incorporates texts and images to “decipher (解码) the brain”, Ariel Goldstein from Princeton University, US, told Science magazine. In the future, scientists hope that the technology can be used to record imagined thoughts and dreams or allow people to understand how differently other animals perceive reality. 1．What do we know about Stable Diffusion? A．It can help train people’s minds. B．It can fully present people’s dreams. C．It was created to draw people’s brains. D．It can produce pictures from people’s thoughts. 2．How did the AI system learn about human brain activities? A．By analyzing descriptions of human thoughts. B．By studying patterns of blood flow in the brain. C．By recording brain scans from four participants. D．By examining 10,000 photos about human brains. 3．How did the AI system improve the quality of the final images? A．By integrating additional brain scans. B．By adjusting various visual elements. C．By employing a “text-to-image” model. D．By inputting text from human observers. 4．What is the potential use of this technology according to scientists? A．To improve text-to-image models. B．To create more accurate brain scans. C．To learn how animals view the world. D．To help people interpret their dreams. 5．What makes the best title for the passage? A．The Research on AI System B．The Power of Science Magazine C．The Forms of Dreams D．The Study for Brain Activities 人工智能的局限性 Passage 1 （24-25高三上·天津·期中）It seems you can ask artificial intelligence (AI) programs anything, and they’ll give you an answer back. So, can they tell jokes? This question was raised when UK comedian Karen Hobbs performed in late June. Instead of the usual jokes she created, Hobbs delivered jokes written by ChatGPT. The process made her very tense. Hobbs said that when she asked ChatGPT for a joke, immediately, what it provided was a man joking about being impatient with his shopping-obsessed girlfriend. Even when she asked to switch to the woman’s perspective, the joke still featured a shopping-obsessed girlfriend, just told from the first-person perspective. It’s no surprise that many jokes come from a male perspective in online materials, as males dominate the field of comedy. The BBC commented that AI storytelling can only reproduce information that already exists in some form, although it can produce some never-before-seen combinations of ideas. “One way that AI can tell jokes is to do what any 5-year-old does — repeat a successful joke they have heard or try to make an obvious variation of it,” said Les Carr, a professor of web science. Nonetheless, a good joke in real life can always adjust to the audience’s feedback, which often leads comedians to improvise (即兴创作) . This, unfortunately, is beyond the current capabilities of AI. For instance, in improvise, there is no space for planning; the comedian must rely only on their instinctive reaction to the audience. “A well-done stand-up bit can lead the audience through a funny story all the way to a hilarious punchline (令人捧腹的妙语),” Michael Ryan, a student researching AI’s impact on comedy, explained to the BBC． “The whole time the comedian knew exactly where he or she was going with the joke and brought the audience there.” However, this could all change. Research is already ongoing with the goal of giving AI a greater understanding of the world around it, which makes the future of Al tokes still uncertain. 1．How did Hobbs feel about the process of telling jokes created by ChatGPT? A．Nervous. B．Ashamed. C．Relaxed. D．Delighted. 2．Why do many jokes come from a male perspective in online materials? A．Women are usually ignored by AI. B．Women don’t like telling jokes. C．Men write most of the jokes. D．Men are more humorous than women. 3．What is AI unable to do with jokes? A．Repeat jokes. B．Adapt jokes. C．Improvise jokes. D．Combine jokes. 4．What is the author’s attitude toward the future of AI-created jokes? A．Uncaring. B．Skeptical. C．Disapproving D．Hopeful. 5．What can we learn from the passage? A．AI has surpassed human comedians in creating original jokes. B．AI’s joke-telling ability still need improving in the future. C．AI is unable to create jokes because it lacks a sense of humor. D．The future of AI in comedy is certain to replace human comedians. Passage 2 （23-24高三下·天津·阶段练习）The Internet has completely changed the workplace over the past three decades. Artificial Intelligence is now all set to do the same, and businesses that don’t take advantage of the technology risk being left behind. Global tech giants like Amazon have been leading the change, and businesses of all sizes are now using the technology for employing and managing their staff. Among them is L’Oreal. With about a million applicants for roughly 15,000 new positions each year, the company is using AI to hire. “We really wanted to save time and focus more on quality, diversity and candidate experience. And AI solutions were the best way to go faster on these challenges,” said Eva Azoulay, global vice-president of L’Oreal’s Human Resources Department. The company uses Mya, a chatbot, to save employers’ time during the first stage of the process. It handles routine questions from candidates, and checks details such as availability and visa requirements. Should candidates make it to the next round, they’ll run into Seedlink, an AI software that scores applicants based on their answers to open-ended interview questions. These scores don’t replace human judgment, said Azoulay, but they do exclude candidates who might not seem like obvious choices. Early results have been promising. For one internship program, where 12,000 people apply for about 80 spots, employers claim they saved 200 hours of time while hiring the most diverse group to date. Other businesses have gone beyond employment and are using AI to help manage employees. Some UK firms have started using Isaak, a system designed by the London-based company StatusToday, to track how many hours staff spend online and the number of emails they receive. London real estate agent JBrown has been using this system since March. CEO James Brown said it helps the firm understand employees’ habits and prevent them from overworking. “It enables us to solve bottleneck problems and relieve overburdened employees,” he said. Despite these examples of good practice, there is still a long way for AI to reach its full potential, and the technology comes with risks. Another AI danger could be its impact on jobs through automation. McKinsey predicts AI could add $13 trillion to the global economy by 2030, with early adopters doubling their cash flow over that period. But the demand for repetitive (重复的) or digitally-unskilled jobs could drop by around 10%, the consulting firm said in a 2018 report. 1．What can we learn about AI technology from Paragraph 1? A．It causes a great problem in workplace. B．It will become a necessary part of business. C．It requires businesses to invest much money. D．It will replace the Internet in the future. 2．L’Oreal uses AI in its hiring process to _____. A．pick out the most suitable candidates directly B．come up with more questions unlimitedly C．improve the company’s hiring efficiency D．save money by replacing human judgment 3．What’s the meaning of the underlined word “exclude” in the 5th paragraph? A．Prepare. B．Consider. C．Remove. D．Include. 4．Firms with the system Isaak can _____. A．prevent their employees from surfing the Internet B．force their employees to form good working habits C．monitor the contents of all their employees’ emails D．help their employees avoid being overstressed at work 5．What is the main idea of the passage? A．What AI will bring to the workplace. B．Why AI could be good for the workplace. C．How businesses can prepare for an AI future. D．How to use AI to improve workplace efficiency. Passage 3 （23-24高三上·天津北辰·阶段练习）Security experts say artificial intelligence (AI) systems used by businesses can make serious, costly mistakes. One way to avoid such mistakes is for companies to employ humans to closely watch AI. The AI system — the credit scoring service — provided by Fair Isaac Corporation (FICO) is used by about two-thirds of the world’s largest banks to help make lending decisions. The system is also used to identify possible cases of credit fraud (欺诈). FICO officials said that one of their company’s AI systems misidentified a large number of credit card fraud cases. The COVID-19 pandemic had caused a large increase in online shopping. The AI system considered the rise in online shopping to be the result of fraudulent activity. As a result, the AI system told banks to deny millions of purchase attempts from online buyers. The incident happened just as people were hurrying to buy products that were in short supply in stores. But in the end, very few buyers had their purchase requests denied. This is because a group of experts the company employs to observe or monitor its AI system recognized the false fraud identifications. The workers made temporary adjustments and succeeded in shunning an AI-ordered block on spending. FICO says the expert team is quickly informed about any unusual buying activity that the AI systems might misidentify. But these kinds of corporate teams are not that common. Last year, FICO and the business advisory company McKinsey Company carried out separate studies on the subject. They found that most organizations involved in the study were not closely watching their AI-based programs. Experts say AI systems mainly make mistakes when real-world situations differ from the situations used in creating the intelligence. In FICO’s case, it said its software expected more in-person shopping than online shopping. This led the system to identify a greater share of financial activity as problematic. Seasonal differences, data-quality changes or extremely unusual events — such as the pandemic — can lead to a series of bad AI predictions. Aleksander Madry is the director of the MIT Center for Deployable Machine Learning. He said the pandemic must have been a “wake-up call” for businesses not closely monitoring their AI systems. “That’s what really stops us currently from this dream of AI revolutionizing everything,” Madry added. 1．Why does the author mention FICO in Paragraph 2? A．To illustrate the main business of the company B．To show the common phenomenon of fraudulent activity C．To give an example of AI systems’ problems in businesses D．To demonstrate the role of AI systems in the financial industry 2．What does the underlined word “shunning” in Paragraph 3 mean? A．Putting off B．Keeping away from C．Making use of D．Giving away 3．What might be the major cause of Al systems’ problems? A．The changes in real situations beyond AI’s database. B．The ignorance of data quality in creating AI systems. C．The increase in credit fraud cases during the pandemic. D．The functional conflicts between different kinds of software. 4．What’s Aleksander Madry’s attitude towards monitoring AI systems? A．Critical B．Unclear C．Suspicious D．Favorable 5．What is the best title for the text? A．Should AI systems be used in banks? B．Do AI systems perform better than humans? C．Should companies trust AI systems completely? D．Can we realize the dream of AI revolutionizing everything? Passage 4 （22-23高三下·天津南开·阶段练习）ChatGPT is a new AI system that sounds so human in conversations that it could host its own radio programs. Reading between its instantly generated, perfectly grammatical lines, people see different visions of the future. Without doubt, ChatGPT is impressive. Some compare the emergence of ChatGPT to the impact of the iPhone, but that doesn’t do it justice. ChatGPT, as well as the generative AI that will follow and outsmart it, is more disruptive. And yet, that doesn’t necessarily mean the end of the world is upon us. On the contrary, ChatGPT, I would argue, might serve to make us more aware of our irreplaceable human qualities. Take the creative act, writing in particular, as an example. If you want it to, the AI-powered chatbot always produces something because it has the whole world of online data to draw from. But unlike us; it lacks the consciousness. Thinking is hard, critical thinking even harder, and ChatGPT isn’t good at either. It just restates what has already been said; it is one big recycling machine. There is another obvious limitation of ChatGPT. Philosopher Harry Frankfurt once claimed: the difference between a bullshitter (胡说八道的人) and a liar is that the liar knows what the truth is but decides to take the opposite direction; a bullshitter, however, has no regard for the truth at all. The AI scholar Gary Marcus applies this distinction to ChatGPT. He believes that we have reached a critical point where “the price of bullshit reaches zero and people who want to spread misinformation, either politically or just to make a profit, start doing that plentifully”. Unfortunately, ChatGPT will reproduce misinformation from any of its input sources — it is not an intelligent system that tries to balance or weight different perspectives. In this sense, everything that ChatGPT writes is bullshit. This is why the so-called AIQ is critical. It is actually an extension and a measurement of our human IQ: our overall knowledge of AI tools, our mastery of clues, and our ethical awareness. ChatGPT is going to change everything and nothing. Creativity, imagination and ethics — these will all remain unique human domains. It is the AI’s very limitations that will make us appreciate our own. 1．What can we learn about ChatGPT? A．It helps generate an artificial voice. B．It provides instructions on writing skills. C．It generates natural language responses. D．It offers a service for language learning. 2．What does the underlined word “disruptive” in paragraph 2 probably mean? A．Evil. B．Reliable. C．Profitable. D．Revolutionary. 3．What’s the writing purpose of Paragraph 3? A．To show the differences between humans and AI. B．To describe the limitations of human consciousness. C．To prove ChatGPT might make humans more aware of their irreplaceable human qualities. D．To explain why ChatGPT isn’t a big recycling machine. 4．Why does the author consider ChatGPT as a bullshit generator? A．It makes up lies constantly. B．It can’t tell right from wrong. C．It often makes unfair judgement. D．It always takes a neutral standpoint. 5．Which can be a suitable title for the text? A．ChatGPT Makes Us Human B．ChatGPT Is Causing Panic Now C．ChatGPT: Treat It Like a Toy, Not a Tool D．ChatGPT: Why It Is Bound to Generate Bullshit 人们对人工智能的态度 Passage 1 （24-25高三上·天津·期末）I’ve considered that fire is our first technology. Its profound and trans-formative impact is still powerful. Prometheus (a god of fire in ancient Greek stories) stole fire from the gods to benefit humanity and disrupted (打乱) the original order to challenge authority. Artificial intelligence (AI) emerges as a modern Prometheus. AI has, in a sense, “stolen” knowledge or rather, made accessible vast amounts of information and insights previously beyond our reach. This act has paved the way for new human achievements, indicating an era of exceptional cognitive (认知的) expansion and controversy. On the one hand, AI has made knowledge widely accessible, broken down barriers to information access, and fueled innovation in every field. AI’s contribution to human progress is obvious. In the meantime, AI’s “theft” of knowledge comes with its own set of challenges and ethical (道德的) dilemmas. The disruption it brings can be seen in job losses, privacy concerns, and the unsettling changes it brings to societal norms. The very act of AI obtaining and processing vast amounts of data raises questions about security and the ethical use of information. So AI challenges us to rethink our relationship with technology and its role in society. As we cast a light at the crossroads of technology and humanity, it is urgent to balance the flame of AI with a commitment to responsible management. In doing so, we can make the most of the full potential of this Promethean gift, ensuring that it serves as a lighthouse of progress and enlightenment, rather than a source of unchecked disruption for you and me. Despite these challenges, AI’s role in our modern society is eminent. It’s similar to that of a partner in human creativity and problem-solving. It takes on the laborious tasks of data analysis and pattern recognition, freeing human minds so that they can engage in higher-order thinking and creative pursuits. This cooperation between humans and AI is fostering a new age of innovation — a cognitive age — where the boundaries of what can be achieved are constantly being redefined. 1．AI is compared to Prometheus in Paragraph 1 to stress that AI_________. A．helps people to succeed easily B．stands up for ordinary people C．brings forth earthshaking changes D．challenges traditional knowledge 2．What is primarily implied about AI in Paragraph 2? A．It is a double-edged sword. B．It is the driving force behind creativity. C．Its cognitive ability needs improvement. D．Its advantages outweigh the potential drawbacks. 3．What is a must in order to maximize AI’s beneficial impact? A．Adopting an open and tolerant mindset. B．Carrying out effective regulation and management. C．Ensuring cross-cultural cooperation in AI development. D．Prioritizing ethical considerations over technological advances. 4．What does the underlined word “eminent” in the last paragraph mean? A．Significant. B．Traditional. C．Conflicting. D．Complicated. 5．What is the author’s general attitude towards AI’s role in our society? A．Concerned. B．Pessimistic. C．Indifferent. D．Positive. Passage 2 （23-24高三上·天津南开·期末）Is it possible to make machines think like humans? This is one question in research in the field of Artificial Intelligence, or “AI” To think like a human, it involves feelings, morality, hopes, and dreams. Humans are capable of making decisions by themselves without input from others, and they can learn from experience. In addition, humans are able to create things from their own inspiration for their own pleasure. The question is whether machines can ever really think in all these aspects. The idea of a computer that is all-powerful and can think and make decisions for itself terrifies many people. In some films, AI even decides that all humans must be killed. Many people are also concerned that AI will be used to replace humans. Yet others are comforted by the thought of AI. Maybe with AI. robots can be made to replace loved ones who have died. Today, the creation of an all-powerful computer capable of human thought is still in progress. We have companion (陪伴) robots, and it is likely that they will grow more important in our lives as time goes by. However, the need for such robots does not seem to be as critical as our need for intelligent machines that can solve problems by learning from their observations and experience.   For example, AI is already being used with great success in Internet search engines and as a personal assistant in our smartphones. It learns from our habits to help us find what we want and like. AI is also used to diagnose (诊断) problems and suggest solutions. Researchers are also experimenting with the use of AI in driverless cars which can sense their surroundings and decide on the best way to reach a destination. It is also being used in education, particularly in online courses. Furthermore, industries that use robots for repetitive or dangerous work often turn to AI to manage these electronic workers. In addition, virtual assistants are also being used in some homes today. In the near future, it is likely that AI will be used to manage smart homes and handle an even greater variety of tasks such as setting out favourite clothes and helping us with our personal grooming (梳妆). The AI of today has already proven its superiority to humans in many areas. Do you think that we can create a robot that thinks like a human or one that is even more intelligent? If so, is there truly a difference between AI and a human? 1．What can we learn from Paragraphs 1 and 2? A．We already have all-powerful computers with the power of human thought. B．One day the AI will decide on its own to wipe out humans. C．Robots with AI have the potential to replace our loved ones. D．In contrast, humans need artificial intelligence machines to help us solve problems rather than companion robots. 2．What does the underlined part in Paragraph 3 probably mean? A． AI will eventually be used to manage workers. B．AI is often used to manage robots on production lines. C．Industrial robots are completely different from electronic workers. D．Industrial robots often turn to AI. 3．What could be inferred from Paragraph 3 about AI? A．Smartphones already have AI technology. B． AI technology for driverless cars is already mature. C． AI has been being used to mange smart homes. D．Artificial intelligence has proved its superiority over humanity in all fields. 4．What is the author’s attitude towards AI? A．Short-sighted. B．Wait-and-see. C．Optimistic. D．Skeptical. 5．What makes the best title for the passage? A． AI Thinks Like Humans B．AI Robots C．AI and Human Beings D．The Future of AI Passage 3 （23-24高三上·天津南开·阶段练习）When it comes to work, workers, and jobs, much of the concern of the modern age boils down to the fear that we’re witnessing the final stage of the game, and that there will be nowhere for humans to withdraw as machines take over the last few tasks. The most recent example comes from the use of artificial intelligence (AI). Is there no area of human experience that can’t be replaced by AI? And if not, what is left for humans to do except the tasks involved in taking care of the machines? At the heart of this concern is our desire for good jobs that make the most of workers’ natural abilities and where the work provides the worker with motivation and work-life balance. More importantly, good jobs support workers in learning by doing——and, in so doing, deliver benefits on three levels:to the worker,who gains in personal development and job satisfaction; to the organization, which reforms as staff find new problems to solve and opportunities to pursue; and to the community as a whole, which harvests the economic benefits of hosting positive organizations and workers. This is what makes good jobs productive for the organization, as well as engaging and fulfilling for the worker. Does the ongoing advance of AI threaten to get rid of all the learning, creativity,and meaning that make a job a good job? Certainly, some have blamed technology for just such an outcome. Headlines today often express concern over technological innovation resulting in bad jobs for humans, or even the complete disappearance of certain professions. Some fear that further technology advancement in the workplace will result in jobs where employees are being asked to work in split times or for longer periods over more days. The problem here isn’t the technology; rather, it’s the way the technology is used —and, more than that, the way people think about using it. 1．What is the passage mainly about? A．How to take care of the AI machines. B．The relations between workers and jobs. C．The human’s fears in the age of AI. D．Human experiences were replaced by AI. 2．What jobs are supposed to be good according to the passage? A．Those that mainly have the economic benefits. B．The ones that offer chances to progress professionally. C．Those that make full use of workers’ motivation. D．The ones that don’t bring new problems to workers. 3．How can good jobs bring about benefits? A．By encouraging the workers to learn while doing. B．By providing the organization with pursuing opportunities. C．By solving new problems among the workers. D．By balancing work and job hours in the community. 4．How do news media respond to the development of AI? A．Negative. B．Supportive. C．Optimistic. D．Unknown. 5．What will the author probably talk about next? A．The future jobs humans will do. B．The application of AI in the future. C．The ways that humans do jobs. D．The proper attitude towards AI. Passage 4 （2023·天津南开·二模）We are encountering real-world examples of how AI can harm human relations. As digital assistants such as Alexa or Siri become popular, we become accustomed to talking to them as though they were alive. Writing in these pages several years ago, Judith Shulevitz described how some of us are starting to treat them as friends and therapists. Shulevitz herself says she confesses (忏悔) things to Google Assistant that she wouldn’t tell her husband. If we grow more comfortable talking to our devices about our secrets, what happens to our human marriages and friendships? Designers and programmers typically create devices whose responses make us feel better — but may not help us be self-reflective or think over painful truths. As AI goes deeper into our lives, we must face the possibility that it will prevent our emotions and deep human connects. Besides, we will fight with some other challenges. The age of driverless cars, after all, is upon us. These vehicles promise to considerably reduce the exhaustion and distraction that put human drivers in danger, thus preventing accidents. But what other effects might they have on people? Driving is a very modern kind of social interaction, requiring high levels of cooperation. I worry that driverless cars, by taking away from us an occasion to exercise this ability, could contribute to its decline. Not only will these vehicles be programmed to take over driving duties and hence to remove from humans the power to make moral judgments (for example, about which pedestrian to hit when a crash is unavoidable), they will also affect humans with whom they’ve had no direct contact. For instance, drivers who have steered awhile alongside an autonomous vehicle traveling at a steady, changeless speed might drive less attentively, thus increasing their likelihood of accidents once they’ve moved to a part of the highway occupied only by human drivers. Alternatively, experience may reveal that driving alongside autonomous vehicles travelling in perfect accordance with (按照) traffic laws actually improves human performance. Either way, we should be careful to launch new forms of AI without first taking such unexpected social effects into account. We must apply the same effort that we apply to the hardware and software that make self-driving cars possible to managing AI’s potential effects on those outside the car. After all, we install brake lights on the back of your car not just, or even primarily, for your benefit, but for the sake of the people behind you. 1．What can be inferred about human relationships from the first paragraph? A．AI will lead to distant inter-personal relationships. B．We will feel comfortable speaking to others online. C．AI will enable people to communicate more with others. D．We will be more self-reflective in interaction thanks to AI. 2．In Paragraph 2, the phrase “its decline” refers to the decline in ________. A．drivers’ interaction with the cars B．drivers’ exhaustion and distraction C．our ability to cooperate with others while driving D．our ability to deal with emergencies while driving 3．According to the passage, which of the following statements is true of driverless cars? A．They may be better at making more judgments than human drivers. B．They need to vary their speed to make contact with human drivers. C．They may make human drivers in other cars drive more safely. D．They need to force human drivers to concentrate in the car. 4．Which of the following is the writer most likely to agree with? A．Brake lights on the back of our car are installed mainly to warn us of danger. B．We should figure out how new technology affects people before developing it. C．We can launch new forms of AI without thinking of unexpected social effects. D．More efforts should be made to advance the hardware and software of driverless cars. 5．What’s the author’s purpose of writing this passage? A．To present the challenges brought by AI. B．To explain the reason why AI may harm human relations. C．To put forward how to solve the unexpected effect of AI. D．To discuss the advantages and disadvantages of driverless cars. 【同源语篇阅读】 【材料来源：The New York Times February 6, 2025】 Why DeepSeek Could Change What Silicon Valley Believes About AI The artificial intelligence breakthrough that is sending shock waves through stock markets, spooking Silicon Valley giants, and generating breathless takes about the end of America's technological dominance arrived with an unassuming, wonky title: "Incentivizing Reasoning Capability in LLMs via Reinforcement Learning." The 22-page paper, released last week by a scrappy Chinese AI start-up called DeepSeek, didn't immediately set off alarm bells. It took a few days for researchers to digest the paper's claims, and the implications of what it described. The company had created a new AI model called DeepSeek-R1, built by a team of researchers who claimed to have used a modest number of second-rate AI chips to match the performance of leading American AI Models at a fraction of the cost. DeepSeek said it had done this by using clever engineering to substitute for raw computing horsepower. And it had done it in China, a country many experts thought was in a distant second place in the global AI race. Some industry watchers initially reacted to DeepSeek's breakthrough with disbelief. Surely, they thought, DeepSeek had cheated to achieve R1's results, or fudged their numbers to make their model look more impressive than it was. Maybe R1 was actually just a clever re-skinning of American AI models that didn't represent much in the way of real progress. Eventually, as more people dug into the details of DeepSeek-R1 - which, unlike most leading AI models, was released as open-source software, allowing outsiders to examine its inner workings more closely - their skepticism morphed into worry. And late last week, when lots of Americans started to use DeepSeek's models for themselves, and the DeepSeek mobile app hit the number one spot on Apple's App Store, it tipped into full-blown panic. Based on conversations I've had with industry insiders, and a week's worth of experts poking around and testing the paper's findings for themselves, it appears to be throwing into question several major assumptions the American tech industry has been making. The first is the assumption that in order to build cutting-edge AI models, you need to spend huge amounts of money on powerful chips and data centers. It's hard to overstate how foundational this dogma has become. Companies like Microsoft, Meta and Google have already spent tens of billions of dollars building out the infrastructure they thought was needed to build and run next-generation AI models. They plan to spend tens of billions more - or, in the case of OpenAI, as much as $500 billion through a joint venture with Oracle and SoftBank that was announced last week. DeepSeek appears to have spent a small fraction of that building R1. The obvious conclusion to draw is not that American tech giants are wasting their money. It's still expensive to run powerful AI models once they're trained, and there are reasons to think that spending hundreds of billions of dollars will still make sense for companies like OpenAI and Google, which can afford to pay dearly to stay at the head of the pack. 【译文欣赏】 为什么DeepSeek可能改变硅谷关于AI的认知 一项人工智能突破在股市掀起冲击波、令硅谷巨头感到恐慌、引发了关于美国技术主导地位终结的热议，这项突破却以一个低调、学究气的标题出现：“通过强化学习激励大语言模型的推理能力”。 这份22页的论文由一家结构松散、名为DeepSeek的中国AI初创公司在上周发布，当时论文并没有立即敲响警钟。 研究人员花了几天时间来消化论文的主张，以及所描述内容的可能影响。 该公司创造了一个名为DeepSeek-R1的AI新模型，构建模型的研究团队声称，他们用数量不多的二流AI芯片、以极低的成本就达到了堪与美国一流AI公司相媲美的性能。 DeepSeek表示，它是通过用巧妙的工程技术来替代原始算力而做到的。 而且它是在中国做到了这一点，许多专家认为中国在全球AI竞赛中处于远远落后的第二位。 一些行业观察家最初对DeepSeek的突破表示怀疑。 他们认为，DeepSeek为了达到R1的结果肯定作了弊，或者篡改了数据，让模型看起来比实际更厉害。 或许R1实际上只是对美国AI模型进行了巧妙的换皮包装，并不代表真正取得了多少进步。 最终，随着越来越多的人深入研究DeepSeek-R1的细节（与大多数领先的AI模型不同，它是作为开源软件发布的，让外部人员能更仔细地审查其内部工作原理），他们的怀疑变成了担忧。 上周晚些时候，当许多美国人开始亲自使用DeepSeek的模型，当DeepSeek移动端应用程序在苹果应用商店排名第一时，这个模型引发了全面的恐慌。 根据我与业内人士的交谈，以及一周以来专家们的探索和对论文结果的亲自测试，这个模型似乎对美国科技行业一直做出的几个主要假设提出了质疑。 第一个假设是，为了构建最尖端的AI模型，你需要在强大的芯片和数据中心上花费巨额资金。 这个教条的根深蒂固怎么夸大都不为过。 微软、Meta、谷歌之类的公司已经花费了数百亿美元来建造他们认为构建和运行下一代AI模型所需的基础设施。 他们计划还要再投入数百亿美元，或者像OpenAI的情况，上周宣布通过与甲骨文和软银的合资企业，再投入多达5000亿美元。 DeepSeek建造R1似乎只花费了这些金额的极小一部分。 可以得出的明显结论并不是美国科技巨头在浪费金钱。 一旦经过训练，运行强大的AI模型仍然花费不菲，而且有理由认为，对于OpenAI和谷歌这样的公司来说，花费数千亿美元仍然是有道理的，因为这些公司有能力付出高昂的代价来保持行业领先地位。 【词汇积累】 source n. 来源；发源地；原始资料 v. 从...获得 impressive adj. 给人深刻印象的；令人钦佩的 inner adj. 内心的；内部的；里面的 n. 里面 technological adj. 科技的；工艺的 disbelief n. 不相信；怀疑 joint adj. 联合的；共同的；连带的；合资的 n. 关节；接头；接合处；接缝；夜总会 v. 贴合；连接 initially adv. 最初 rate n. 比率；速度；价格；费用；等级 v. 认为；估价；定等级；值得；怒斥；责骂 generation n. 一代（人）；一代人的时间；（产品发展的）代；产生 obvious adj.明显的;显著的;平淡无奇的;自明 panic n. 惊慌；恐慌 v. 使…惊慌；感到恐慌 adj. 恐慌的；惊慌的 distant adj. 疏远的；遥远的；远房的；陌生的；远行的 assumption n. 假定；设想；担任(职责等)；假装 raw adj. 未加工的；生的；原始的；湿冷的 n. 触到某人的痛处 breakthrough n. 突破，突破性进展；关键问题的解决 make sense 讲得通； 有意义； 言之有理 leading adj. 领导的；指导的；主要的；在前的 【词汇拓展】 venture n. 冒险(事业)；风险；投机；企业 v. 冒险；敢于；冒昧地说 infrastructure n. 基础；基础设施 digest n. 摘要 vt. 消化；理解 vi. 消化 breathless adj. 喘不过气来的 horsepower n. 马力(功率单位) dearly adv. 深深地；真挚地；昂贵地 overstate v. 夸大，对…言过其实，夸大的叙述 a fraction of 一小部分, 少许 engineering n. 工程学；工程 dominance n. 支配；控制；统治；优势；【生】优势度，显性 full-blown adj. 盛开的；成熟的；全面的 substitute for 用…代替， （使）代替;替换;抵换 dogma n. 教条；教义；信条 skepticism n. 怀疑论；怀疑态度；怀疑主义 ( 1 )原创精品资源学科网独家享有版权，侵权必究！ 学科网（北京）股份有限公司 $$ 热点02 DeepSeek人工智能AI 在科技浪潮汹涌澎湃的今天，人工智能AI正以前所未有的速度重塑着我们的世界。在这股浪潮中，DeepSeek如同一颗璀璨的新星，以其卓越的智能解决方案，正引领着未来的方向。DeepSeek不仅仅是一款AI模型，它是智慧与创新的结晶，以其独特的魅力在人工智能领域独树一帜。让我们一同探索DeepSeek的奥秘，见证它如何以其卓越的智能，为我们的生活带来革命性的变化，开启智能科技的新纪元。 （建议用时：60分钟） 人工智能对人类的影响 Passage 1 （2024·天津十二区重点校·一模）假设你是晨光中学高三学生李津。你有一位英国好友Jim，他告诉你他最近经常使用ChatGPT（一款人工智能的语言类软件）来完成他的学习任务，比如写演讲稿和实验报告。他问你是否愿意使用这类人工智能来完成学习任务，你写给他一封E-mail，主要内容包括： 1. 你是否愿意让人工智能替你完成学习任务及其原因。 2. 你在生活中是怎样利用高科技产品的。 3. 祝愿Jim学业进步。 注意：1. 词数100左右：2. 开头和结尾已给出，不计入总词数。 词汇提示： 人工智能：AI (Artificial Intelligence)  高科技产品：high-tech products Dear Jim, ____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ Yours, Li Jin 【答案】Dear Jim, I heard that you often use ChatGPT to assist in your learning tasks, and I would like to share my thoughts on this matter. Personally, I am not inclined to use artificial intelligence to help do my learning tasks. While its efficiency and accuracy are indeed impressive, we can gain a deeper understanding of knowledge only by learning independently. In my daily life, I actively embrace high-tech products as well. They make it possible for me to access information, learn new things, and stay connected with friends anytime, anywhere. At the meantime, I don’t let them control my life by setting a fixed time for using them. Finally, I wish you all the best in your academic journey. I am confident that with your dedication and hard work, you will achieve excellent results. Yours, Li Jin 【导语】本篇书面表达属于应用文，要求考生给Jim写一封电子邮件，告诉他你是否愿意让人工智能替自己完成学习任务及其原因、在生活中是怎样利用高科技产品的、并祝愿Jim学业进步。 【详解】1. 词汇积累 帮助：assist→help 倾向于：be inclined to do...→tend to do.../have a tendency to do... 独立：independently→on one’s own 与……保持联系：stay connected with→keep in touch with 2. 句式拓展 合并简单句 原句：In my daily life, I actively embrace high-tech products as well. They make it possible for me to access information, learn new things, and stay connected with friends anytime, anywhere. 拓展句：In my daily life, I actively embrace high-tech products as well, which make it possible for me to access information, learn new things, and stay connected with friends anytime, anywhere. 【点睛】【高分句型1】While its efficiency and accuracy are indeed impressive, we can gain a deeper understanding of knowledge only by learning independently. (使用了while引导让步状语从句、动名词作宾语) 【高分句型2】They make it possible for me to access information, learn new things, and stay connected with friends anytime, anywhere. (使用了make it possible for sb. to do...结构) Passage 2 （2023·天津·模拟预测）假设你是振华中学高三学生李明。你与交换生James要共同完成一篇项目学习的论文，他建议用AI完成初稿。请你用英语给他写一封电子邮件，内容包括： 1. 委婉拒绝并说明理由； 2. 你的建议及计划。 注意：1. 词数100左右； 2. 开头和结尾已给出，不计入总词数。 提示词语：项目学习 project-based learning Dear James, ____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ Yours, Li Ming 【答案】Dear James, Thanks for your suggestion on using AI to finish the first draft of our paper. But after further consideration, I’m afraid it’s not a wise idea. Firstly, it goes against the purpose of project-based learning, which is to cultivate critical thinking and teamwork. Moreover, AI is unlikely to accurately present our ideas, because it hasn’t “got involved in” our project. So, I suggest we write the paper ourselves. We can start by brainstorming together to create an outline after generating some original and creative ideas. Then we can work independently on different parts, and finally work together to finalize our paper. I believe we can do a better job than AI. Hope you can understand. Expecting our distinctive draft. Yours, Li Ming 【导语】本篇书面表达属于应用文。题目要求考生给交换生James写邮件，委婉地拒绝他的用AI完成项目论文初稿的建议，并陈述理由，然后提出自己的建议和论文写作计划。 【详解】1. 词汇积累 完成：finish→complete, accomplish, finalize 考虑：consideration→thought, deliberation, reflection, contemplation 首先：firstly→to begin with, in the first place, at first, at the beginning 培养：cultivate→develop, nurture, foster 此外：moreover→besides, in addition, additionally, furthermore 不太可能的：unlikely→improbable, questionable, beyond belief 呈现：present→show, display, exhibit, demonstrate 产生：generate→produce, create 2. 句式拓展 合并简单句 原句：We can start by brainstorming together to create an outline after generating some original and creative ideas. Then we can work independently on different parts, and finally work together to finalize our paper. 拓展句：We can start by brainstorming together to create an outline after generating some original and creative ideas, after which we can work independently on different parts, and finally work together to finalize our paper. 【点睛】【高分句型1】But after further consideration, I’m afraid it’s not a wise idea.（运用了连接词that引导名词性从句作afraid的宾语并省略了that） 【高分句型2】Firstly, it goes against the purpose of project-based learning, which is to cultivate critical thinking and teamwork.（运用了关系代词which引导非限制性定语从句) 【高分句型3】Moreover, AI is unlikely to accurately present our ideas, because it hasn’t “got involved in” our project.（运用了从属连词because引导原因状语从句） Passage 3 （24-25高三上·天津·阶段练习）A study conducted by Cornell University examined how the use of AI in conversations impacts the way people express themselves and view each other. The researchers have found people have more efficient conversations, use more positive language and perceive each other more positively when using an AI-enabled chat tool. However the group also found that when participants think their partner is using more AI-suggested responses, they consider that partner as less cooperative. “I was surprised to find people tend to evaluate you more negatively simply because they suspect you’re using AI to help compose text, regardless of whether you actually are,” said Jess Hohenstein, the lead researcher. “This illustrates the continuous overall doubt that people seem to have around AI.” For their first experiment, participants were asked to talk about a policy issue and assigned to one of three conditions: both participants can use smart replies; only one participant can use smart replies; or neither participant can use smart replies. Researchers found that using smart replies increased communication efficiency, positive emotional language and positive evaluations by communication partners. On average, smart replies accounted for 14.3% of sent messages. But participants whose partners suspected of responding with smart replies were evaluated more negatively than those who were thought to have typed their own responses, consistent with common assumptions about the negative implications of AI. “While AI might be able to help you write,” Hohenstein said, “it’s altering your language in ways you might not expect, especially by making you sound more positive. This suggests that by, using text-generating AI, you’re giving up some of your-own personal voice.” Malte Jung, an associate professor, said, “What we observe in this study is the impact that AI has on social dynamics and some of the unintended consequences that could result from integrating AI in social contexts. This suggests that whoever controls the algorithm (算法) may have influence on people’s interactions, language and insights into each other.” 6．Which of the following statements does Jess Hohenstein agree with? A．AI always expresses in ways you expect. B．Algorithm will never influence people’s insights. C．Trust can be affected by using AI in conversation. D．You will regain your voice by using AI in conversation. 7．How do the researchers draw their conclusion? A．By analyzing figures. B．By performing a survey. C．By conducting experiments. D．By making comparisons. 8．How will a person feel about suspecting his partner’s using smart replies? A．Nervous. B．Uncomfortable. C．Excited. D．Puzzled. 9．According to the passage, what are the unexpected consequences of using text- generating AI? A．Your language remains unchanged. B．It doesn’t alter speakers’ insights into each other. C．It makes the speaker sound more negative. D．You are losing some of your personal style. 5．What is the passage primarily about? A．Methods of using AI in conversations. B．Efficiency of using AI in conversations. C．Convenience of using AI in conversations. D．Impacts of using AI in conversations. 【答案】6．C 7．C 8．B 9．D 5．D 【导语】本文是一篇说明文。文章通过介绍康奈尔大学的一项研究，探讨了人工智能在对话中的使用如何影响人们的表达方式和相互看法。 6．推理判断题。根据文章第四段““I was surprised to find people tend to evaluate you more negatively simply because they suspect you’re using AI to help compose text, regardless of whether you actually are,” said Jess Hohenstein, the lead researcher.(首席研究员Jess Hohenstein说：“让我惊讶的是，人们往往会因为怀疑你使用AI来帮助编写文本而对你产生更负面的评价，无论你是否真的在使用。”)”可知，这表明了Jess Hohenstein认为在对话中使用AI会影响信任。故选C。 7．细节理解题。根据文章第五段“For their first experiment, participants were asked to talk about a policy issue and assigned to one of three conditions: both participants can use smart replies; only one participant can use smart replies; or neither participant can use smart replies.(在第一项实验中，参与者被要求讨论一个政策问题，并被分配到三种情况之一：双方都可以使用智能回复；只有一方可以使用智能回复；或者双方都不能使用智能回复。)”可知，研究者是通过实验来得出结论的。故选C。 8．细节理解题。根据文章第六段“But participants whose partners suspected of responding with smart replies were evaluated more negatively than those who were thought to have typed their own responses, consistent with common assumptions about the negative implications of AI.(但是，当参与者怀疑对方使用智能回复进行回复时，他们对对方的评价要比认为对方是自己打字回复时的评价更负面，这与人们对AI负面影响的普遍看法一致。)”可知，人们会因为怀疑伙伴使用智能回复而感到不舒服或产生负面评价。故选B。 9．细节理解题。根据文章第七段““While AI might be able to help you write,” Hohenstein said, “it’s altering your language in ways you might not expect, especially by making you sound more positive. This suggests that by, using text-generating AI, you’re giving up some of your-own personal voice.”(Hohenstein说：“虽然AI可能有助于写作，但它会以你可能意想不到的方式改变你的语言，尤其是让你听起来更加积极。这表明，通过使用文本生成AI，你正在放弃自己的一部分个人声音。”)”可知，使用文本生成AI，人们会放弃一些自己的个人声音或风格。故选D。 5．主旨大意题。根据文章第一段“A study conducted by Cornell University examined how the use of AI in conversations impacts the way people express themselves and view each other.(康奈尔大学进行的一项研究探讨了人工智能（AI）在对话中的使用如何影响人们的表达方式和相互看法。)”以及全文内容可知，文章通过介绍康奈尔大学的研究，探讨了AI在对话中的使用如何影响人们的表达方式和相互看法，以及可能导致的信任问题和个人风格的丧失等。故选D。 Passage 4 （2024·天津九校·二模）After astonishing breakthroughs in artificial intelligence, many people worry that they will be replaced by AI. Global Google searches for “is my job safe?” have doubled in recent months, as people fear that they will be replaced with large language models (LLMs). In a recent paper Tyna Eloundou of OpenAI and colleagues say that “around 80% of the US workforce could have at least 10% of their work tasks affected by the introduction of LLMs”. Economists, however, tend to enjoy making predictions about automation more than they enjoy testing them. In the early 2010s many of them loudly predicted that robots would kill jobs by the millions, only to fall silent when employment rates across the rich world rose to all-time highs. Few of the doom-mongers (末日论者) have a good explanation for why countries with the highest rates of tech usage around the globe, such as Japan, Singapore and the Republic of Korea, consistently have among the lowest rates of unemployment. Here we introduce our first attempt at tracking AI’s impact on jobs. Using American data on employment by occupation, we single out white-collar workers. White-collar roles are thought to be especially sensitive to generative AI, which is becoming ever better at logical reasoning and creativity. However, there is as yet little evidence of an AI hit to employment. In the spring of 2020 white-collar jobs rose, as many people in service occupations lost their job at the start of the Covid-19 pandemic. And in the past year the white-collar share of employment in professions thought to be at risk from generative AI has risen by 0.5%. Of course, it’s still early days for generative AI. Few firms yet use generative Al tools at large scale, so the impact on jobs could merely be delayed. Another possibility, however, is that these new technologies will end up destroying only a small number of roles. While Al may be efficient at some tasks, it may be less good at others, such as management and working out what others need. AI could even have a positive effect on jobs. If workers using it become more efficient, profits at their company could rise which would then allow bosses to ramp up hiring. A recent survey by Experis, an IT-recruitment firm, points to this possibility. More than half of Britain’s employers expect AI technologies to have a positive impact on the number of their staff over the next two years, it finds. To see how it all shakes out, we will publish updates to this analysis every few months. But for now, a job apocalypse seems a long way off. 1．In terms of artificial intelligence, people worry about ______. A．how to avoid being replaced by machines B．how to keep up with the trend of technology C．how to master new technological skills D．how to earn higher wages by using technology 2．Economists’ predictions about automation show that ______. A．machines prove to deal a blow to employment B．technological breakthroughs bring an economic boom C．economic theories don’t work well in practice D．it is difficult to clarify the impact of machines on jobs 3．According to the text, since the spring of 2020, white-collar jobs ______. A．have become more diverse B．have decreased in number C．have been threatened by service jobs D．have not suffered noticeable losses 4．How may artificial intelligence influence employment? A．It may cause mass unemployment. B．It may create new jobs for people. C．It may enable employers to hire more. D．It may decrease people’s salaries. 5．What could be the best title for the passage? A．Boy Cries Wolf: Generative Al is not yet killing jobs B．Batten Down the Hatches: Generative AI will replace most of human jobs C．Generative AI: The greatest threat to white-collar workers. D．Generative AI: The greatest invention of the 20th century 【答案】1．A 2．D 3．D 4．C 5．A 【导语】本文是一篇说明文。这篇文章主要介绍了人工智能对就业的影响。 1．细节理解题。根据第一段“After astonishing breakthroughs in artificial intelligence, many people worry that they will be replaced by AI. Global Google searches for “is my job safe?” have doubled in recent months, as people fear that they will be replaced with large language models (LLMs). In a recent paper Tyna Eloundou of OpenAI and colleagues say that   “around 80% of the US workforce could have at least 10% of their work tasks affected by the introduction of LLMs”.(在人工智能取得惊人突破之后，许多人担心他们将被人工智能取代。谷歌搜索“我的工作安全吗？”的数量在最近几个月翻了一番，因为人们担心它们将被大型语言模型(LLMs)所取代。在最近的一篇论文中，OpenAI的Tyna Eloundou和他的同事说，“大约80%的美国劳动力至少有10%的工作任务可能会受到引入人工智能的影响。”)”可知，许多人担心他们会被AI取代，因此他们担心如何避免被机器取代。故选A项。 2．推理判断题。根据第二段“Economists, however, tend to enjoy making predictions about automation more than they enjoy testing them. In the early 2010s many of them loudly predicted that robots would kill jobs by the millions, only to fall silent when employment rates across the rich world rose to all-time highs.(然而，经济学家更喜欢预测自动化，而不是测试自动化。在2010年代初，他们中的许多人大声预测机器人将夺走数百万人的工作，但当发达国家的就业率升至历史最高水平时，他们却沉默了)”可知，经济学家喜欢预测自动化但不喜欢测试，之前预测机器人会消灭数百万工作机会，但实际上就业率却上升了，这表明机器对工作岗位的影响很难明确。故选D项。 3．细节理解题。根据第五段“White-collar roles are thought to be especially sensitive to generative AI, which is becoming ever better at logical reasoning and creativity. However, there is as yet little evidence of an AI hit to employment. In the spring of 2020 white-collar jobs rose, as many people in service occupations lost their job at the start of the Covid-19 pandemic. And in the past year the white-collar share of employment in professions thought to be at risk from generative AI has risen by 0.5%.(白领角色被认为对生成式人工智能特别敏感，因为生成式人工智能在逻辑推理和创造力方面正变得越来越好。然而，目前几乎没有证据表明人工智能对就业造成了冲击。2020年春季，白领岗位增加，因为许多服务行业的人在covid -19大流行开始时失去了工作。在过去的一年里，被认为受到人工智能威胁的职业中，白领所占的比例上升了0.5%)”可知，尽管有人预测生成式人工智能将对白领工作产生影响，但实际上在2020年春季以来，白领工作并没有受到明显的冲击，甚至在被认为受到生成式人工智能威胁的职业中，白领工作的比例还上升了0.5%。因此，可以得出结论，白领工作自2020年春季以来并未遭受明显损失。故选D项。 4．细节理解题。根据倒数第三段“AI could even have a positive effect on jobs. If workers using it become more efficient, profits at their company could rise which would then allow bosses to ram p up hiring.(人工智能甚至可能对就业产生积极影响。如果使用它的员工变得更有效率，公司的利润就会上升，这将使老板们能够加快招聘)”以及倒数第二段“A recent survey by Experis, an IT-recruitment firm, points to this possibility. More than half of Britain’s employers expect AI technologies to have a positive impact on the number of their staff over the next two years, it finds.(Experis是一家 IT 招聘公司，其最近的一项重新认证调查显示，有这种可能性。调查显示，超过半数的英国雇主预计人工智能技术在未来两年内会对员工数量产生积极影响)”可知，如果工人使用AI后效率提高，公司利润可能上升，这将使老板能够增加招聘，因此人工智能可能会促使雇主雇佣更多的人。故选C项。 5．主旨大意题。根据第一段“After astonishing breakthroughs in artificial intelligence, many people worry that they will be replaced by AI. Global Google searches for “is my job safe?” have doubled in recent months, as people fear that they will be replaced with large language models (LLMs). In a recent paper Tyna Eloundou of OpenAI and colleagues say that   “around 80% of the US workforce could have at least 10% of their work tasks affected by the introduction of LLMs”.(在人工智能取得惊人突破之后，许多人担心他们将被人工智能取代。谷歌搜索“我的工作安全吗?”的数量在最近几个月翻了一番，因为人们担心它们将被大型语言模型(LLMs)所取代。在最近的一篇论文中，OpenAI的Tyna Eloundou和他的同事说，“大约80%的美国劳动力至少有10%的工作任务可能会受到引入人工智能的影响。”)”以及最后一段“To see how it all shakes out, we will publish updates to this analysis every few months. But for now, a job apocalypse seems a long way off.(为了了解结果如何，我们将每隔几个月更新一次分析。但就目前而言，就业末日似乎还很遥远)”以及纵观全文可知，本文主要讨论了生成式人工智能对工作岗位的影响，但目前还没有明确的证据表明它会大规模取代人类工作，因此最好的标题应该是A项“Boy Cries Wolf: Generative Al is not yet killing jobs(狼来了：生成式人工智能尚未扼杀工作岗位)”是本文最好的标题。故选A项。 人工智能工具介绍 Passage 1 （24-25高三上·天津南开·阶段练习）Imagine this. You need an image of a balloon for a work presentation and turn to an Al text-to-image generator, like Midjourney or DALL-E, to create a suitable image. You enter the prompt (提示词) “red balloon against a blue sky” but the generator returns an image of an egg instead. What’s going on? The generator you’re using may have been “poisoned”. What does this mean? Text-to-image generators work by being trained on large datasets that include millions or billions of images. Some of the generators have been trained by — indiscriminately scraping online images, many of which may be under copyright. This has led to many copyright infringement (侵害) cases where artists have accused big tech companies of stealing and profiting from their work. This is also where the idea of “poison” comes in. Researchers who want to empower individual artists have recently created a tool named “Nightshade” to fight back against unauthorised image scraping. The tool works by slightly altering an image’s pixels (像素) in a way that confuses the computer vision system but leaves the image unaltered to a human’s eyes. If an organization then scrapes one of these images to train a future AI model, its data pool becomes “poisoned”. This can result in mistaken learning, which makes the generator return unintended results. As in our earlier example, a balloon might become an egg. The higher the number of “poisoned” images in the training data, the greater the impact. Because of how generative AI works, the damage from “poisoned” images also affects related prompt keywords. For example, if a “poisoned” image of a Picasso work is used in training data, prompt results for masterpieces from other artists can also be affected. Possibly, tools like Nightshade can be abused by some users to intentionally upload “poisoned” images in order to confuse AI generators. But the Nightshade’s developer hopes the tool will make big tech companies more respectful of copyright. It does challenge a common belief among computer scientists that data found online can be used for any purpose they see fit. Human rights activists, for example, have been concerned for some time about the indiscriminate use of machine vision in wider society. This concern is particularly serious concerning facial recognition. There is a clear connection between facial recognition cases and data poisoning, as both relate to larger questions around technological governance. It may be better to see data poisoning as an innovative solution to the denial of some fundamental human rights. 1．The first paragraph is used to ____________. A．report a piece of news B．inform an academic accident C．arouse readers’ interest D．introduce some AI generators 2．The underlined word ‘scraping’ (para. 2) is closest in meaning to ____________. A．facilitating B．collecting C．damaging D．polishing 3．According to the passage, adding poisoned data might ____________. A．increase the accuracy of returned information B．cause users to forget the prompt key words C．interfere with the training of generative AI D．discriminate against great masterpieces 4．What can be inferred from the last two paragraphs? A．Data poisoning is somehow justified to direct attention to human rights. B．Computer scientists have learned to respect the copyright of most artists. C．Nightshade is being abused by human rights activists to recognize faces. D．The issue of technological governance has aroused the lawyers’ interest. 5．Which of the following might be the best title of the passage? A．Data Poisoning: Government Empowering Citizens to Protect Themselves B．Data Poisoning: Addressing Facial Recognition Issues Among Artists C．Data Poisoning: Risks and Rewards of Generative AI Data Training D．Data Poisoning: Restricting Innovation or Empowering Artists 【答案】1．C 2．B 3．C 4．A 5．D 【导语】本文是一篇说明文。文章介绍了AI文本生成图像工具的工作原理、可能存在的版权问题以及由此引出的“投毒”概念，还探讨了相关工具如Nightshade的用途和可能带来的影响，以及这一领域与人权、技术治理等议题的关联。 1．推理判断题。根据文章第一段“Imagine this. You need an image of a balloon for a work presentation and turn to an Al text-to-image generator, like Midjourney or DALL-E, to create a suitable image. You enter the prompt(提示词) “red balloon against a blue sky” but the generator returns an image of an egg instead.(想象一下这样的场景。你需要一张气球的图片用于工作汇报，于是转向使用AI文本生成图像的工具，比如Midjourney或DALL-E，来创建一张合适的图片。你输入提示词“蓝天下的红色气球”，但生成器返回的却是一张鸡蛋的图片。)”可知，第一段描述了一个具体的场景：你需要一张气球的图片用于工作汇报，但使用AI文本生成图像工具时，输入了“蓝天下的红色气球”的提示词，却得到了一张鸡蛋的图片。这个场景既有趣又引人好奇，目的是吸引读者的注意力，让他们对接下来要讨论的内容产生兴趣。故选C。 2．词义猜测题。根据划线单词上文“Text-to-image generators work by being trained on large datasets that include millions or billions of images.(文本生成图像的工具是通过在包含数百万或数十亿张图片的大型数据集上进行训练来工作的。)”可知，文本生成图像工具需要分析网上大量的照片，由此可知，文本生成图像工具得从网上收集图片。引出，划线单词和“收集”意思相似。选项A“facilitating (促进)”；选项B“collecting (收集)”；选项C“damaging (损害)”；选项D“polishing (润色)”。故选B。 3．细节理解题。根据文章第三段“The tool works by slightly altering an image’s pixels(像素)in a way that confuses the computer vision system but leaves the image unaltered to a human’s eyes. If an organization then scrapes one of these images to train a future AI model, its data pool becomes “poisoned”. This can result in mistaken learning, which makes the generator return unintended results.(这个工具的工作原理是轻微地改变图像的像素，以一种让计算机视觉系统感到困惑但人类肉眼却看不出变化的方式。如果一个组织随后抓取这些图像来训练未来的AI模型，那么它的数据集就变得“有毒”了。这可能会导致错误的学习，使得生成器返回非预期的结果。)”可知，“Poisoned data”是指被故意添加进数据集以干扰AI模型训练的数据。这种数据旨在破坏模型的性能，使其无法正常工作或产生错误的结果。故选C。 4．推理判断题。根据文章倒数第二段“Possibly, tools like Nightshade can be abused by some users to intentionally upload “poisoned” images in order to confuse AI generators. But the Nightshade’s developer hopes the tool will make big tech companies more respectful of copyright.(可能有些用户会滥用像Nightshade这样的工具，故意上传“有毒”图像以混淆AI生成器。但Nightshade的开发者希望这个工具能让大型科技公司更加尊重版权。)”可知，；再根据文章最后一段“Human rights activists, for example, have been concerned for some time about the indiscriminate use of machine vision in wider society. This concern is particularly serious concerning facial recognition. There is a clear connection between facial recognition cases and data poisoning, as both relate to larger questions around technological governance. It may be better to see data poisoning as an innovative solution to the denial of some fundamental human rights. (例如，人权活动家们已经有一段时间对机器视觉在社会中的滥用表示担忧。这种担忧在面部识别方面尤为严重。面部识别案件和数据中毒之间存在明显的联系，因为两者都涉及到更广泛的技术治理问题。将数据中毒视为对某些基本人权剥夺的创新解决方案，或许是一个更好的视角。)”可知，最后两段提到，人权活动人士一直担心滥用机器视觉，特别是面部识别，并且面部识别案件与数据中毒之间存在联系。作者建议，将数据中毒视为剥夺某些基本人权的创新解决方案可能是一个更好的观点。这意味着，数据中毒可以被视为引起人们对人权问题的注意的一种方式，从而证明在某些情况下使用数据中毒是合理的。故选A。 5．主旨大意题。根据文章第一段“You enter the prompt(提示词) “red balloon against a blue sky” but the generator returns an image of an egg instead.(你输入提示词“蓝天下的红色气球”，但生成器返回的却是一张鸡蛋的图片。)”、第二段“The generator you’re using may have been “poisoned”. What does this mean? Text-to-image generators work by being trained on large datasets that include millions or billions of images. Some of the generators have been trained by —indiscriminately scraping online images, many of which may be under copyright. This has led to many copyright infringement(侵害)cases where artists have accused big tech companies of stealing and profiting from their work.(你正在使用的生成器可能已经被“投毒”了。这是什么意思呢？文本生成图像的工具是通过在包含数百万或数十亿张图片的大型数据集上进行训练来工作的。其中一些生成器是通过不加区分地从网上抓取图片来进行训练的，而这些图片中很多可能都受版权保护。这导致了许多版权侵权案件，艺术家们指责大型科技公司窃取他们的作品并从中获利。)”、倒数第二段“Possibly, tools like Nightshade can be abused by some users to intentionally upload “poisoned” images in order to confuse AI generators. But the Nightshade’s developer hopes the tool will make big tech companies more respectful of copyright.(可能有些用户会滥用像Nightshade这样的工具，故意上传“有毒”图像以混淆AI生成器。但Nightshade的开发者希望这个工具能让大型科技公司更加尊重版权。)”以及最后一段“Human rights activists, for example, have been concerned for some time about the indiscriminate use of machine vision in wider society.(例如，人权活动家已经有一段时间对机器视觉在社会中的不加区分使用表示担忧。)”可知，文章主要解释了“中毒”数据的含义，即通过特地工具对图像像素进行细微修改，使得这些图像在训练AI模型时会导致模型学习错误，从而产生非预期的输出，这导致了许多侵犯版权的案件，引起艺术家的不满。选项D“Data Poisoning: Restricting Innovation or Empowering Artists(数据中毒：限制创新还是赋权艺术家)”总结了全文内容，为最佳标题。故选D。 Passage 2 （2024·天津河东·二模）After waking up, you may feel frustrated that you cannot recall the dreams you had last night. Artificial intelligence (AI) may be able to help. Previously, there have been AI models that can turn text into images. They can do this by learning from a large amount of data from both tests and images. This time, researchers from Osaka University in Japan have trained an AI system called Stable Diffusion to re-create images based on people’s brain scans, reported Science magazine. The researchers used an online data set provided by the University of Minnesota, US, which consisted of brain scans from four participants as they each viewed a set of 10,000 photos. The scans were recorded by functional magnetic resonance imaging (FMRI,功能性磁共振成像). The AI then learned about the brain activities by analyzing changes in blood flow shown by the FMRI data - when a part of the brain is activated, more blood will flow to it. It then matched the brain activities with the photos. Through this method, AI learned how human brains would react when seeing different photos. Finally, the researchers tested the AI on additional brain scans from the same participants when they viewed photos of a toy bear, airplane, clock and train. If the person looked at an airplane, for example, the AI would use the brain scan data to create an image of a very blurry (模糊的) airplane. Then, it would turn on the previous “text-to-image” model and improve the quality of the image by feeding itself the keyword “airplane”. The final images were “convincing” with about 80 percent of accuracy, according to the researchers. The new study created a novel approach that incorporates texts and images to “decipher (解码) the brain”, Ariel Goldstein from Princeton University, US, told Science magazine. In the future, scientists hope that the technology can be used to record imagined thoughts and dreams or allow people to understand how differently other animals perceive reality. 1．What do we know about Stable Diffusion? A．It can help train people’s minds. B．It can fully present people’s dreams. C．It was created to draw people’s brains. D．It can produce pictures from people’s thoughts. 2．How did the AI system learn about human brain activities? A．By analyzing descriptions of human thoughts. B．By studying patterns of blood flow in the brain. C．By recording brain scans from four participants. D．By examining 10,000 photos about human brains. 3．How did the AI system improve the quality of the final images? A．By integrating additional brain scans. B．By adjusting various visual elements. C．By employing a “text-to-image” model. D．By inputting text from human observers. 4．What is the potential use of this technology according to scientists? A．To improve text-to-image models. B．To create more accurate brain scans. C．To learn how animals view the world. D．To help people interpret their dreams. 5．What makes the best title for the passage? A．The Research on AI System B．The Power of Science Magazine C．The Forms of Dreams D．The Study for Brain Activities 【答案】1．D 2．B 3．C 4．C 5．A 【导语】这是一篇说明文。主要介绍了日本大阪大学的研究人员训练了一个名为“Stable Diffusion”的人工智能系统，该系统可以根据人们的脑部扫描重新创建图像，具有现实意义。 1．细节理解题。根据第二段中“This time, researchers from Osaka University in Japan have trained an AI system called Stable Diffusion to re-create images based on people’s brain scans, reported Science magazine.”（据《科学》杂志报道，这一次，日本大阪大学的研究人员训练了一个名为“Stable Diffusion”的人工智能系统，该系统可以根据人们的脑部扫描重新创建图像。）可知，Stable Diffusion可以从人们的思想中生成图像。故选D项。 2．细节理解题。根据第四段“The AI then learned about the brain activities by analyzing changes in blood flow shown by the FMRI data - when a part of the brain is activated, more blood will flow to it. It then matched the brain activities with the photos. Through this method, AI learned how human brains would react when seeing different photos.”（然后，人工智能通过分析FMRI数据显示的血流变化来了解大脑活动——当大脑的一部分被激活时，更多的血液会流向它。然后将大脑活动与照片进行匹配。通过这种方法，人工智能了解了人类大脑在看到不同照片时的反应。）可知，人工智能系统通过研究大脑血液流动的模式来了解人类大脑活动的。故选B项。 3．细节理解题。根据倒数第二段“Finally, the researchers tested the AI on additional brain scans from the same participants when they viewed photos of a toy bear, airplane, clock and train. If the person looked at an airplane, for example, the AI would use the brain scan data to create an image of a very blurry (模糊的) airplane. Then, it would turn on the previous “text-to-image” model and improve the quality of the image by feeding itself the keyword “airplane”. The final images were “convincing” with about 80 percent of accuracy, according to the researchers.”（最后，研究人员在这些参与者观看玩具熊、飞机、时钟和火车的照片时，对人工智能进行了额外的脑部扫描。例如，如果一个人看着一架飞机，人工智能将使用大脑扫描数据创建一个非常模糊的飞机图像。然后，它会打开之前的“文本到图像”模型，并通过输入关键字“飞机”来提高图像的质量。根据研究人员的说法，最终的图像“令人信服”，准确率约为80%。）可知，人工智能系统通过采用“文本到图像”的模式来提高最终图像的质量。故选C项。 4．推理判断题。根据最后一段“The new study created a novel approach that incorporates texts and images to “decipher (解码) the brain”, Ariel Goldstein from Princeton University, US, told Science magazine. In the future, scientists hope that the technology can be used to record imagined thoughts and dreams or allow people to understand how differently other animals perceive reality.”（美国普林斯顿大学的阿里尔·戈尔茨坦在接受《科学》杂志采访时表示，这项新研究创造了一种结合文本和图像来“破译大脑”的新方法。在未来，科学家们希望这项技术可以用来记录想象中的想法和梦想，或者让人们了解其他动物对现实的感知有多么不同。）可推知，科学家认为这项技术的潜在用途是人们了解动物是如何感知现实的。故选C项。 5．主旨大意题。根据文章大意以及第二段中“This time, researchers from Osaka University in Japan have trained an AI system called Stable Diffusion to re-create images based on people’s brain scans, reported Science magazine.”（据《科学》杂志报道，这一次，日本大阪大学的研究人员训练了一个名为“Stable Diffusion”的人工智能系统，该系统可以根据人们的脑部扫描重新创建图像。）可知，文章主要讲的是一个名为“Stable Diffusion”的人工智能系统，该系统可以根据人们的脑部扫描重新创建图像。由此可知，文章最好的标题是“人工智能系统研究”。故选A项。 人工智能的局限性 Passage 1 （24-25高三上·天津·期中）It seems you can ask artificial intelligence (AI) programs anything, and they’ll give you an answer back. So, can they tell jokes? This question was raised when UK comedian Karen Hobbs performed in late June. Instead of the usual jokes she created, Hobbs delivered jokes written by ChatGPT. The process made her very tense. Hobbs said that when she asked ChatGPT for a joke, immediately, what it provided was a man joking about being impatient with his shopping-obsessed girlfriend. Even when she asked to switch to the woman’s perspective, the joke still featured a shopping-obsessed girlfriend, just told from the first-person perspective. It’s no surprise that many jokes come from a male perspective in online materials, as males dominate the field of comedy. The BBC commented that AI storytelling can only reproduce information that already exists in some form, although it can produce some never-before-seen combinations of ideas. “One way that AI can tell jokes is to do what any 5-year-old does — repeat a successful joke they have heard or try to make an obvious variation of it,” said Les Carr, a professor of web science. Nonetheless, a good joke in real life can always adjust to the audience’s feedback, which often leads comedians to improvise (即兴创作) . This, unfortunately, is beyond the current capabilities of AI. For instance, in improvise, there is no space for planning; the comedian must rely only on their instinctive reaction to the audience. “A well-done stand-up bit can lead the audience through a funny story all the way to a hilarious punchline (令人捧腹的妙语),” Michael Ryan, a student researching AI’s impact on comedy, explained to the BBC． “The whole time the comedian knew exactly where he or she was going with the joke and brought the audience there.” However, this could all change. Research is already ongoing with the goal of giving AI a greater understanding of the world around it, which makes the future of Al tokes still uncertain. 1．How did Hobbs feel about the process of telling jokes created by ChatGPT? A．Nervous. B．Ashamed. C．Relaxed. D．Delighted. 2．Why do many jokes come from a male perspective in online materials? A．Women are usually ignored by AI. B．Women don’t like telling jokes. C．Men write most of the jokes. D．Men are more humorous than women. 3．What is AI unable to do with jokes? A．Repeat jokes. B．Adapt jokes. C．Improvise jokes. D．Combine jokes. 4．What is the author’s attitude toward the future of AI-created jokes? A．Uncaring. B．Skeptical. C．Disapproving D．Hopeful. 5．What can we learn from the passage? A．AI has surpassed human comedians in creating original jokes. B．AI’s joke-telling ability still need improving in the future. C．AI is unable to create jokes because it lacks a sense of humor. D．The future of AI in comedy is certain to replace human comedians. 【答案】1．A 2．C 3．C 4．B 5．B 【导语】本文是一篇说明文。文章分析了人工智能创作笑话的局限性。 1．细节理解题。根据第一段“It seems you can ask artificial intelligence (AI) programs anything, and they’ll give you an answer back. So, can they tell jokes? This question was raised when UK comedian Karen Hobbs performed in late June. Instead of the usual jokes she created, Hobbs delivered jokes written by ChatGPT. The process made her very tense.(似乎你可以问人工智能程序任何问题，它们会给你答案。那么，他们会讲笑话吗？这个问题是在6月底英国喜剧演员凯伦·霍布斯的表演中提出的。霍布斯没有像往常一样讲她创作的笑话，而是讲了ChatGPT写的笑话。这个过程使她非常紧张)”可知，霍布斯在讲ChatGPT创造的讲笑话时感到很紧张。故选A项。 2．细节理解题。根据第二段的最后一句“It’s no surprise that many jokes come from a male perspective in online materials, as males dominate the field of comedy.(由于男性主导了喜剧领域，网络素材中许多笑话都来自男性视角，这并不奇怪)”可知，很多网络段子都是从男性角度切入的是因为大部分笑话都是男人写的。故选C项。 3．细节理解题。根据第四段的句子“Nonetheless, a good joke in real life can always adjust to the audience’s feedback, which often leads comedians to improvise (即兴创作) . This, unfortunately, is beyond the current capabilities of AI.(尽管如此，现实生活中一个好的笑话总是可以根据观众的反馈进行调整，这通常会导致喜剧演员即兴发挥。不幸的是，这超出了人工智能目前的能力)”可知，讲笑话时，人工智能不能即兴创作。故选C项。 4．推理判断题。根据最后一段“However, this could all change. Research is already ongoing with the goal of giving AI a greater understanding of the world around it, which makes the future of Al tokes still uncertain.(然而，这一切都可能改变。研究已经在进行，目的是让人工智能更好地了解周围的世界，这使得人工智能创作笑话的未来仍然不确定)”可知，作者认为工智能创作笑话的未来仍然不确定，所以作者对人工智能创作笑话的未来持怀疑的态度。故选B项。 5．推理判断题。根据第四段的句子“Nonetheless, a good joke in real life can always adjust to the audience’s feedback, which often leads comedians to improvise (即兴创作) . This, unfortunately, is beyond the current capabilities of AI.(尽管如此，现实生活中一个好的笑话总是可以根据观众的反馈进行调整，这通常会导致喜剧演员即兴发挥。不幸的是，这超出了人工智能目前的能力。)”和文章最后一段“However, this could all change. Research is already ongoing with the goal of giving AI a greater understanding of the world around it, which makes the future of Al tokes still uncertain.(然而，这一切都可能改变。研究已经在进行，目的是让人工智能更好地了解周围的世界，这使得人工智能创作笑话的未来仍然不确定。)”可知，未来，人工智能的讲笑话能力仍有待提高。故选B项。 Passage 2 （23-24高三下·天津·阶段练习）The Internet has completely changed the workplace over the past three decades. Artificial Intelligence is now all set to do the same, and businesses that don’t take advantage of the technology risk being left behind. Global tech giants like Amazon have been leading the change, and businesses of all sizes are now using the technology for employing and managing their staff. Among them is L’Oreal. With about a million applicants for roughly 15,000 new positions each year, the company is using AI to hire. “We really wanted to save time and focus more on quality, diversity and candidate experience. And AI solutions were the best way to go faster on these challenges,” said Eva Azoulay, global vice-president of L’Oreal’s Human Resources Department. The company uses Mya, a chatbot, to save employers’ time during the first stage of the process. It handles routine questions from candidates, and checks details such as availability and visa requirements. Should candidates make it to the next round, they’ll run into Seedlink, an AI software that scores applicants based on their answers to open-ended interview questions. These scores don’t replace human judgment, said Azoulay, but they do exclude candidates who might not seem like obvious choices. Early results have been promising. For one internship program, where 12,000 people apply for about 80 spots, employers claim they saved 200 hours of time while hiring the most diverse group to date. Other businesses have gone beyond employment and are using AI to help manage employees. Some UK firms have started using Isaak, a system designed by the London-based company StatusToday, to track how many hours staff spend online and the number of emails they receive. London real estate agent JBrown has been using this system since March. CEO James Brown said it helps the firm understand employees’ habits and prevent them from overworking. “It enables us to solve bottleneck problems and relieve overburdened employees,” he said. Despite these examples of good practice, there is still a long way for AI to reach its full potential, and the technology comes with risks. Another AI danger could be its impact on jobs through automation. McKinsey predicts AI could add $13 trillion to the global economy by 2030, with early adopters doubling their cash flow over that period. But the demand for repetitive (重复的) or digitally-unskilled jobs could drop by around 10%, the consulting firm said in a 2018 report. 1．What can we learn about AI technology from Paragraph 1? A．It causes a great problem in workplace. B．It will become a necessary part of business. C．It requires businesses to invest much money. D．It will replace the Internet in the future. 2．L’Oreal uses AI in its hiring process to _____. A．pick out the most suitable candidates directly B．come up with more questions unlimitedly C．improve the company’s hiring efficiency D．save money by replacing human judgment 3．What’s the meaning of the underlined word “exclude” in the 5th paragraph? A．Prepare. B．Consider. C．Remove. D．Include. 4．Firms with the system Isaak can _____. A．prevent their employees from surfing the Internet B．force their employees to form good working habits C．monitor the contents of all their employees’ emails D．help their employees avoid being overstressed at work 5．What is the main idea of the passage? A．What AI will bring to the workplace. B．Why AI could be good for the workplace. C．How businesses can prepare for an AI future. D．How to use AI to improve workplace efficiency. 【答案】1．B 2．C 3．C 4．D 5．A 【导语】本文是一篇说明文。文章介绍了人工智能将如何改变工作场所，一些公司正在利用人工智能技术来招聘和管理员工，还有一些公司还开始使用人工智能系统来帮助跟踪员工的工作时间和行为。尽管人工智能有很多优点，但也伴随着一些风险，其中之一是导致部分重复性或非数字技术工作需求的下降。 1．推理判断题。根据第一段“The Internet has completely changed the workplace over the past three decades. Artificial Intelligence is now all set to do the same, and businesses that don’t take advantage of the technology risk being left behind. (在过去的三十年里，互联网彻底改变了工作场所。人工智能现在也在做同样的事情，不利用这项技术的企业有被甩在后面的风险)”可知，人工智能技术将改变工作场所，如果企业不用它，则有可能被淘汰。由此可知，人工智能将成为商业的必要组成部分。故选B项。 2．细节理解题。根据第四段中“We really wanted to save time and focus more on quality, diversity and candidate experience. And AI solutions were the best way to go faster on these challenges (我们真的想节省时间，把更多的精力放在质量、多样性和候选人经验上。而人工智能解决方案是应对这些挑战的最佳方式)”可知，欧莱雅在招聘过程中使用人工智能来提高公司的招聘效率。故选C项。 3．词句猜测题。根据画线词的上文“These scores don’t replace human judgment (这些分数并不能取代人类的判断)”可知，此处是介绍人工智能在招聘中所打的分数的用处，说明它是辅助性的，再结合表示转折的“but”和下文“candidates who might not seem like obvious choices (那些看起来不太合适的候选人)”可推知，但它的确可以起到作用，帮助“排除”明显不合适的候选人，画线词意思应该与Remove“排除”相近。故选C项。 4．细节理解题。根据倒数第三段中“Some UK firms have started using Isaak, a system designed by the London-based company StatusToday, to track how many hours staff spend online and the number of emails they receive. (一些英国公司已经开始使用由伦敦公司StatusToday设计的Isaak系统来跟踪员工上网的时间和他们收到的电子邮件数量。)”和“CEO James Brown said it helps the firm understand employees’ habits and prevent them from overworking. (首席执行官James Brown说，这有助于公司了解员工的习惯，防止他们过度工作)”可知，有Isaak系统的公司可以帮助员工避免工作压力过大。故选D项。 5．主旨大意题。通读全文，尤其是第一段中“The Internet has completely changed the workplace over the past three decades. Artificial Intelligence is now all set to do the same (在过去的三十年里，互联网彻底改变了工作场所。人工智能现在也在做同样的事情)”可知，文章围绕人工智能将如何改变工作场所展开，介绍了一些公司正在利用人工智能技术来招聘和管理员工，还有一些公司还开始使用人工智能系统来帮助跟踪员工的工作时间和行为。尽管人工智能有很多优点，但也伴随着一些风险，其中之一是导致部分重复性或非数字技术工作需求的下降。故选A项。 Passage 3 （23-24高三上·天津北辰·阶段练习）Security experts say artificial intelligence (AI) systems used by businesses can make serious, costly mistakes. One way to avoid such mistakes is for companies to employ humans to closely watch AI. The AI system — the credit scoring service — provided by Fair Isaac Corporation (FICO) is used by about two-thirds of the world’s largest banks to help make lending decisions. The system is also used to identify possible cases of credit fraud (欺诈). FICO officials said that one of their company’s AI systems misidentified a large number of credit card fraud cases. The COVID-19 pandemic had caused a large increase in online shopping. The AI system considered the rise in online shopping to be the result of fraudulent activity. As a result, the AI system told banks to deny millions of purchase attempts from online buyers. The incident happened just as people were hurrying to buy products that were in short supply in stores. But in the end, very few buyers had their purchase requests denied. This is because a group of experts the company employs to observe or monitor its AI system recognized the false fraud identifications. The workers made temporary adjustments and succeeded in shunning an AI-ordered block on spending. FICO says the expert team is quickly informed about any unusual buying activity that the AI systems might misidentify. But these kinds of corporate teams are not that common. Last year, FICO and the business advisory company McKinsey Company carried out separate studies on the subject. They found that most organizations involved in the study were not closely watching their AI-based programs. Experts say AI systems mainly make mistakes when real-world situations differ from the situations used in creating the intelligence. In FICO’s case, it said its software expected more in-person shopping than online shopping. This led the system to identify a greater share of financial activity as problematic. Seasonal differences, data-quality changes or extremely unusual events — such as the pandemic — can lead to a series of bad AI predictions. Aleksander Madry is the director of the MIT Center for Deployable Machine Learning. He said the pandemic must have been a “wake-up call” for businesses not closely monitoring their AI systems. “That’s what really stops us currently from this dream of AI revolutionizing everything,” Madry added. 1．Why does the author mention FICO in Paragraph 2? A．To illustrate the main business of the company B．To show the common phenomenon of fraudulent activity C．To give an example of AI systems’ problems in businesses D．To demonstrate the role of AI systems in the financial industry 2．What does the underlined word “shunning” in Paragraph 3 mean? A．Putting off B．Keeping away from C．Making use of D．Giving away 3．What might be the major cause of Al systems’ problems? A．The changes in real situations beyond AI’s database. B．The ignorance of data quality in creating AI systems. C．The increase in credit fraud cases during the pandemic. D．The functional conflicts between different kinds of software. 4．What’s Aleksander Madry’s attitude towards monitoring AI systems? A．Critical B．Unclear C．Suspicious D．Favorable 5．What is the best title for the text? A．Should AI systems be used in banks? B．Do AI systems perform better than humans? C．Should companies trust AI systems completely? D．Can we realize the dream of AI revolutionizing everything? 【答案】1．C 2．B 3．A 4．D 5．C 【导语】这是一篇说明文。文章主要讲述，企业使用的人工智能系统可能会犯下严重且代价高昂的错误，所以公司应雇佣人类密切关注人工智能。 1．推理判断题。由文章第二段“The AI system — the credit scoring service — provided by Fair Isaac Corporation (FICO) is used by about two-thirds of the world’s largest banks to help make lending decisions. The system is also used to identify possible cases of credit fraud (欺诈). FICO officials said that one of their company’s AI systems misidentified a large number of credit card fraud cases. The COVID-19 pandemic had caused a large increase in online shopping. The AI system considered the rise in online shopping to be the result of fraudulent activity.(Fair Isaac Corporation（FICO）提供的人工智能系统——信用评分服务——被世界上大约三分之二的最大银行用来帮助做出贷款决策。该系统还用于识别可能的信用欺诈案件。FICO官员表示，他们公司的一个人工智能系统错误识别了大量信用卡欺诈案件。新冠肺炎大流行导致网上购物大幅增加。人工智能系统认为网上购物的兴起是欺诈活动的结果。)”可知，文章提到 FICO是为了举一个人工智能系统在商业中存在问题的例子。故选C项。 2．词句猜测题。由文章第三段“But in the end, very few buyers had their purchase requests denied. This is because a group of experts the company employs to observe or monitor its AI system recognized the false fraud identifications. The workers made temporary adjustments and succeeded in shunning an AI-ordered block on spending.(但最终，很少有买家的购买请求被拒绝。这是因为该公司雇佣了一组专家来观察或监控其人工智能系统，识别出了虚假的欺诈标识。工人们进行了临时调整，成功地 人工智能下令的支出限制。)”可知，这里指工人们成功“避开”了人工智能做出的错误决定。所以猜测shunning是“避开”的意思。故选B项。 3．推理判断题。由文章第五段“Experts say AI systems mainly make mistakes when real-world situations differ from the situations used in creating the intelligence. In FICO’s case, it said its software expected more in-person shopping than online shopping. This led the system to identify a greater share of financial activity as problematic. Seasonal differences, data-quality changes or extremely unusual events — such as the pandemic — can lead to a series of bad AI predictions.(专家表示，当现实世界的情况与创建智能时使用的情况不同时，人工智能系统主要会出错。在FICO的案例中，该公司表示，其软件预计更多的是面对面购物，而不是网上购物。这导致该系统将更大比例的金融活动认定为有问题。季节性差异、数据质量变化或极不寻常的事件——如疫情——都可能导致一系列糟糕的人工智能预测。)”可知，人工智能数据库之外的真实情况的变化是AI系统发生问题的原因。故选A项。 4．推理判断题。由文章最后一段“Aleksander Madry is the director of the MIT Center for Deployable Machine Learning. He said the pandemic must have been a‘wake-up call’ for businesses not closely monitoring their AI systems. ‘That’s what really stops us currently from this dream of AI revolutionizing everything,’ Madry added.(Aleksander Madry是麻省理工学院可部署机器学习中心的主任。他说，对于没有密切监控其人工智能系统的企业来说，这场大流行一定是一个“警钟”。“这是目前真正阻止我们实现人工智能彻底改变一切的梦想的原因。’)”可知，Aleksander Madry是非常支持人工监控人工智能系统。故选D项。 5．主旨大意题。由文章第一段“Security experts say artificial intelligence (AI) systems used by businesses can make serious, costly mistakes. One way to avoid such mistakes is for companies to employ humans to closely watch AI.(安全专家表示，企业使用的人工智能系统可能会犯下严重且代价高昂的错误。避免此类错误的一种方法是公司雇佣人类密切关注人工智能。)”以及后文可知，文章主要讲述，企业使用的人工智能系统可能会犯下严重错误，所以公司应雇佣人类密切关注人工智能。所以C项Should companies trust AI systems completely?(公司应该完全信任人工智能系统吗？)符合语境。故选C项。 Passage 4 （22-23高三下·天津南开·阶段练习）ChatGPT is a new AI system that sounds so human in conversations that it could host its own radio programs. Reading between its instantly generated, perfectly grammatical lines, people see different visions of the future. Without doubt, ChatGPT is impressive. Some compare the emergence of ChatGPT to the impact of the iPhone, but that doesn’t do it justice. ChatGPT, as well as the generative AI that will follow and outsmart it, is more disruptive. And yet, that doesn’t necessarily mean the end of the world is upon us. On the contrary, ChatGPT, I would argue, might serve to make us more aware of our irreplaceable human qualities. Take the creative act, writing in particular, as an example. If you want it to, the AI-powered chatbot always produces something because it has the whole world of online data to draw from. But unlike us; it lacks the consciousness. Thinking is hard, critical thinking even harder, and ChatGPT isn’t good at either. It just restates what has already been said; it is one big recycling machine. There is another obvious limitation of ChatGPT. Philosopher Harry Frankfurt once claimed: the difference between a bullshitter (胡说八道的人) and a liar is that the liar knows what the truth is but decides to take the opposite direction; a bullshitter, however, has no regard for the truth at all. The AI scholar Gary Marcus applies this distinction to ChatGPT. He believes that we have reached a critical point where “the price of bullshit reaches zero and people who want to spread misinformation, either politically or just to make a profit, start doing that plentifully”. Unfortunately, ChatGPT will reproduce misinformation from any of its input sources — it is not an intelligent system that tries to balance or weight different perspectives. In this sense, everything that ChatGPT writes is bullshit. This is why the so-called AIQ is critical. It is actually an extension and a measurement of our human IQ: our overall knowledge of AI tools, our mastery of clues, and our ethical awareness. ChatGPT is going to change everything and nothing. Creativity, imagination and ethics — these will all remain unique human domains. It is the AI’s very limitations that will make us appreciate our own. 1．What can we learn about ChatGPT? A．It helps generate an artificial voice. B．It provides instructions on writing skills. C．It generates natural language responses. D．It offers a service for language learning. 2．What does the underlined word “disruptive” in paragraph 2 probably mean? A．Evil. B．Reliable. C．Profitable. D．Revolutionary. 3．What’s the writing purpose of Paragraph 3? A．To show the differences between humans and AI. B．To describe the limitations of human consciousness. C．To prove ChatGPT might make humans more aware of their irreplaceable human qualities. D．To explain why ChatGPT isn’t a big recycling machine. 4．Why does the author consider ChatGPT as a bullshit generator? A．It makes up lies constantly. B．It can’t tell right from wrong. C．It often makes unfair judgement. D．It always takes a neutral standpoint. 5．Which can be a suitable title for the text? A．ChatGPT Makes Us Human B．ChatGPT Is Causing Panic Now C．ChatGPT: Treat It Like a Toy, Not a Tool D．ChatGPT: Why It Is Bound to Generate Bullshit 【答案】1．C 2．D 3．C 4．B 5．A 【导语】这是一篇说明文。ChatGPT是一种新的人工智能系统，它在对话中听起来非常人性化，甚至可以主持自己的广播节目，但是ChatGPT的局限性让人类更加意识到他们不可替代的人类品质。 1．推理判断题。根据第一段第一句“ChatGPT is a new AI system that sounds so human in conversations that it could host its own radio programs.(ChatGPT是一种新的人工智能系统，它在对话中听起来非常人性化，甚至可以主持自己的广播节目。)”可知，关于ChatGPT我们能了解到它会产生自然的语言反应。故选C。 2．词句猜测题。根据第二段第一句“Some compare the emergence of ChatGPT to the impact of the iPhone, but that doesn’t do it justice.(有些人将ChatGPT的出现与iPhone的影响相提并论，但这并不公平。)”和下文“And yet, that doesn’t necessarily mean the end of the world is upon us. (然而，这并不一定意味着世界末日就要来临。)”可知，ChatGPT，以及紧随其后并超越它的生成式人工智能，更具颠覆性。此处是指ChatGPT的影响更具颠覆性。所以disruptive意为“革命性的，突破性的”。故选D。 3．推理判断题。根据第二段最后一句“On the contrary, ChatGPT, I would argue, might serve to make us more aware of our irreplaceable human qualities.(相反，我认为ChatGPT可能会让我们更多地意识到我们不可替代的人类品质。)”和第三段第一句“Take the creative act, writing in particular, as an example.(以创造性行为，尤其是写作为例。)”可知，第三段的写作目的是为了证明ChatGPT可能会让人类更加意识到他们不可替代的人类品质。故选C。 4．细节理解题。根据第四段最后两句“Unfortunately, ChatGPT will reproduce misinformation from any of its input sources — it is not an intelligent system that tries to balance or weight different perspectives. In this sense, everything that ChatGPT writes is bullshit.(不幸的是，ChatGPT会从任何输入源复制错误信息——它不是一个试图平衡或权衡不同观点的智能系统。从这个意义上说，ChatGPT写的所有东西都是废话。)”可知，作者认为ChatGPT是一个废话生成器是因为它不能分辨是非。故选B。 5．主旨大意题。根据最后一段“Creativity, imagination and ethics — these will all remain unique human domains. It is the AI’s very limitations that will make us appreciate our own.(创造力、想象力和道德——这些都将是人类独有的领域。正是人工智能的局限性会让我们欣赏自己的独特性。)”可知，文章主要是讲述ChatGPT让人类更加意识到他们不可替代的人类品质。因此推断A项“ChatGPT让我们成为人类”为最佳标题。故选A。 人们对人工智能的态度 Passage 1 （24-25高三上·天津·期末）I’ve considered that fire is our first technology. Its profound and trans-formative impact is still powerful. Prometheus (a god of fire in ancient Greek stories) stole fire from the gods to benefit humanity and disrupted (打乱) the original order to challenge authority. Artificial intelligence (AI) emerges as a modern Prometheus. AI has, in a sense, “stolen” knowledge or rather, made accessible vast amounts of information and insights previously beyond our reach. This act has paved the way for new human achievements, indicating an era of exceptional cognitive (认知的) expansion and controversy. On the one hand, AI has made knowledge widely accessible, broken down barriers to information access, and fueled innovation in every field. AI’s contribution to human progress is obvious. In the meantime, AI’s “theft” of knowledge comes with its own set of challenges and ethical (道德的) dilemmas. The disruption it brings can be seen in job losses, privacy concerns, and the unsettling changes it brings to societal norms. The very act of AI obtaining and processing vast amounts of data raises questions about security and the ethical use of information. So AI challenges us to rethink our relationship with technology and its role in society. As we cast a light at the crossroads of technology and humanity, it is urgent to balance the flame of AI with a commitment to responsible management. In doing so, we can make the most of the full potential of this Promethean gift, ensuring that it serves as a lighthouse of progress and enlightenment, rather than a source of unchecked disruption for you and me. Despite these challenges, AI’s role in our modern society is eminent. It’s similar to that of a partner in human creativity and problem-solving. It takes on the laborious tasks of data analysis and pattern recognition, freeing human minds so that they can engage in higher-order thinking and creative pursuits. This cooperation between humans and AI is fostering a new age of innovation — a cognitive age — where the boundaries of what can be achieved are constantly being redefined. 1．AI is compared to Prometheus in Paragraph 1 to stress that AI_________. A．helps people to succeed easily B．stands up for ordinary people C．brings forth earthshaking changes D．challenges traditional knowledge 2．What is primarily implied about AI in Paragraph 2? A．It is a double-edged sword. B．It is the driving force behind creativity. C．Its cognitive ability needs improvement. D．Its advantages outweigh the potential drawbacks. 3．What is a must in order to maximize AI’s beneficial impact? A．Adopting an open and tolerant mindset. B．Carrying out effective regulation and management. C．Ensuring cross-cultural cooperation in AI development. D．Prioritizing ethical considerations over technological advances. 4．What does the underlined word “eminent” in the last paragraph mean? A．Significant. B．Traditional. C．Conflicting. D．Complicated. 5．What is the author’s general attitude towards AI’s role in our society? A．Concerned. B．Pessimistic. C．Indifferent. D．Positive. 【答案】1．C 2．A 3．B 4．A 5．D 【导语】这是一篇说明文。文章主要论述了人工智能所带来的巨大变化以及其优缺点，和为了最大限度地发挥人工智能的有益影响，必须实施有效的监管。 1．细节理解题。根据第一段“Artificial intelligence (AI) emerges as a modern Prometheus. AI has, in a sense, “stolen” knowledge or rather, made accessible vast amounts of information and insights previously beyond our reach. This act has paved the way for new human achievements, indicating an era of exceptional cognitive (认知的) expansion and controversy.(人工智能以现代普罗米修斯的形象出现。从某种意义上说，人工智能“窃取”了知识，或者更确切地说，让我们能够获得大量以前无法获得的信息和见解。这一举动为人类的新成就铺平了道路，标志着一个异常的认知扩张和争议的时代)”可知，将AI比作普罗米修斯，强调AI带来翻天覆地的变化。故选C。 2．推理判断题。根据第二段“On the one hand, AI has made knowledge widely accessible, broken down barriers to information access, and fueled innovation in every field. AI’s contribution to human progress is obvious. In the meantime, AI’s “theft” of knowledge comes with its own set of challenges and ethical (道德的) dilemmas. The disruption it brings can be seen in job losses, privacy concerns, and the unsettling changes it brings to societal norms. The very act of AI obtaining and processing vast amounts of data raises questions about security and the ethical use of information.(一方面，人工智能使知识广泛传播，打破了信息获取的障碍，推动了各个领域的创新。人工智能对人类进步的贡献是显而易见的。与此同时，人工智能对知识的“窃取”也带来了一系列挑战和伦理困境。它带来的破坏可以从失业、隐私问题以及它给社会规范带来的令人不安的变化中看到。人工智能获取和处理大量数据的行为本身就引发了有关信息安全和道德使用的问题)”可知，暗示AI是一把双刃剑。故选A。 3．细节理解题。根据第三段“As we cast a light at the crossroads of technology and humanity, it is urgent to balance the flame of AI with a commitment to responsible management.(当我们在技术和人类的十字路口投下光芒时，迫切需要平衡人工智能的火焰与负责任的管理承诺)”可知，为了最大限度地发挥人工智能的有益影响，必须实施有效的监管。故选B。 4．词句猜测题。根据划线词后文“It’s similar to that of a partner in human creativity and problem-solving. It takes on the laborious tasks of data analysis and pattern recognition, freeing human minds so that they can engage in higher-order thinking and creative pursuits.(它类似于人类创造力和解决问题的伙伴。它承担了繁重的数据分析和模式识别任务，解放了人类的思想，使他们能够从事更高层次的思考和创造性的追求)”可知，人工智能承担了繁重的数据分析和模式识别任务，解放了人类的思想，使他们能够从事更高层次的思考和创造性的追求，说明其在社会中的作用是重要的。故划线词意思是“重要的”。故选A。 5．推理判断题。根据最后一段“Despite these challenges, AI’s role in our modern society is eminent. It’s similar to that of a partner in human creativity and problem-solving. It takes on the laborious tasks of data analysis and pattern recognition, freeing human minds so that they can engage in higher-order thinking and creative pursuits. This cooperation between humans and AI is fostering a new age of innovation — a cognitive age — where the boundaries of what can be achieved are constantly being redefined.(尽管存在这些挑战，人工智能在我们现代社会中的作用是显著的。它类似于人类创造力和解决问题的伙伴。它承担了繁重的数据分析和模式识别任务，解放了人类的思想，使他们能够从事更高层次的思考和创造性的追求。人类和人工智能之间的这种合作正在培育一个创新的新时代——一个认知时代——在这个时代，可以实现的界限正在不断被重新定义)”可知，作者对人工智能在我们社会中的作用的总体态度是积极的。故选D。 Passage 2 （23-24高三上·天津南开·期末）Is it possible to make machines think like humans? This is one question in research in the field of Artificial Intelligence, or “AI” To think like a human, it involves feelings, morality, hopes, and dreams. Humans are capable of making decisions by themselves without input from others, and they can learn from experience. In addition, humans are able to create things from their own inspiration for their own pleasure. The question is whether machines can ever really think in all these aspects. The idea of a computer that is all-powerful and can think and make decisions for itself terrifies many people. In some films, AI even decides that all humans must be killed. Many people are also concerned that AI will be used to replace humans. Yet others are comforted by the thought of AI. Maybe with AI. robots can be made to replace loved ones who have died. Today, the creation of an all-powerful computer capable of human thought is still in progress. We have companion (陪伴) robots, and it is likely that they will grow more important in our lives as time goes by. However, the need for such robots does not seem to be as critical as our need for intelligent machines that can solve problems by learning from their observations and experience.   For example, AI is already being used with great success in Internet search engines and as a personal assistant in our smartphones. It learns from our habits to help us find what we want and like. AI is also used to diagnose (诊断) problems and suggest solutions. Researchers are also experimenting with the use of AI in driverless cars which can sense their surroundings and decide on the best way to reach a destination. It is also being used in education, particularly in online courses. Furthermore, industries that use robots for repetitive or dangerous work often turn to AI to manage these electronic workers. In addition, virtual assistants are also being used in some homes today. In the near future, it is likely that AI will be used to manage smart homes and handle an even greater variety of tasks such as setting out favourite clothes and helping us with our personal grooming (梳妆). The AI of today has already proven its superiority to humans in many areas. Do you think that we can create a robot that thinks like a human or one that is even more intelligent? If so, is there truly a difference between AI and a human? 1．What can we learn from Paragraphs 1 and 2? A．We already have all-powerful computers with the power of human thought. B．One day the AI will decide on its own to wipe out humans. C．Robots with AI have the potential to replace our loved ones. D．In contrast, humans need artificial intelligence machines to help us solve problems rather than companion robots. 2．What does the underlined part in Paragraph 3 probably mean? A． AI will eventually be used to manage workers. B．AI is often used to manage robots on production lines. C．Industrial robots are completely different from electronic workers. D．Industrial robots often turn to AI. 3．What could be inferred from Paragraph 3 about AI? A．Smartphones already have AI technology. B． AI technology for driverless cars is already mature. C． AI has been being used to mange smart homes. D．Artificial intelligence has proved its superiority over humanity in all fields. 4．What is the author’s attitude towards AI? A．Short-sighted. B．Wait-and-see. C．Optimistic. D．Skeptical. 5．What makes the best title for the passage? A． AI Thinks Like Humans B．AI Robots C．AI and Human Beings D．The Future of AI 【答案】1．D 2．B 3．A 4．C 5．C 【导语】这是一篇说明文。文章主要探讨了人工智能与人类之间的关系。 1．细节理解题。由文章第二段中“Yet others are comforted by the thought of AI. Maybe with AI. robots can be made to replace loved ones who have died. Today, the creation of an all-powerful computer capable of human thought is still in progress. We have companion (陪伴) robots, and it is likely that they will grow more important in our lives as time goes by. However, the need for such robots does not seem to be as critical as our need for intelligent machines that can solve problems by learning from their observations and experience. (然而，另一些人对人工智能的想法感到欣慰。也许是人工智能。机器人可以用来代替死去的亲人。今天，制造一台能理解人类思想的全能计算机仍在进行中。我们有伴侣机器人，随着时间的推移，它们很可能会在我们的生活中变得越来越重要。然而，对这种机器人的需求似乎没有我们对智能机器的需求那么重要，智能机器可以通过观察和经验来解决问题。)”可知，相比之下，人类更需要人工智能机器来帮助我们解决问题，而不是陪伴机器人。故选D。 2．词句猜测题。由文章第三段中“industries that use robots for repetitive or dangerous work often turn to AI to manage these electronic workers.”可知，这是一个主从复合句，主句是industries often turn to AI to manage these electronic workers“工业通常转向人工智能来管理这些电子工人”，定语从句是that use robots for repetitive or dangerous work“使用机器人进行重复或危险的工作的”。结合选项：A. AI will eventually be used to manage workers人工智能最终将用于管理工人；B. AI is often used to manage robots on production lines人工智能通常用于管理生产线上的机器人；C. Industrial robots are completely different from electronic workers工业机器人与电子工人完全不同；D.Industrial robots often turn to AI工业机器人经常转向人工智能，可知B选项与划线句意思相近。故选B。 3．推理判断题。由文章第三段中“For example, AI is already being used with great success in Internet search engines and as a personal assistant in our smartphones. It learns from our habits to help us find what we want and like. (例如，人工智能已经在互联网搜索引擎和智能手机的个人助理中获得了巨大的成功。它从我们的习惯中学习，帮助我们找到我们想要和喜欢的东西。)”可知，智能手机已经拥有人工智能技术。故选A。 4．推理判断题。由文章第三段中“The AI of today has already proven its superiority to humans in many areas. (今天的人工智能已经在许多领域证明了它对人类的优越性。)”可知，作者对人工智能的态度是乐观的。故选C。 5．主旨大意题。通读全文，尤其是由文章第一段中“Many people are also concerned that AI will be used to replace humans. (许多人还担心人工智能将被用来取代人类。)”和第三段中“The AI of today has already proven its superiority to humans in many areas. (今天的人工智能已经在许多领域证明了它对人类的优越性。)”可知，本文主要探讨了人工智能与人类之间的关系。故选C。 Passage 3 （23-24高三上·天津南开·阶段练习）When it comes to work, workers, and jobs, much of the concern of the modern age boils down to the fear that we’re witnessing the final stage of the game, and that there will be nowhere for humans to withdraw as machines take over the last few tasks. The most recent example comes from the use of artificial intelligence (AI). Is there no area of human experience that can’t be replaced by AI? And if not, what is left for humans to do except the tasks involved in taking care of the machines? At the heart of this concern is our desire for good jobs that make the most of workers’ natural abilities and where the work provides the worker with motivation and work-life balance. More importantly, good jobs support workers in learning by doing——and, in so doing, deliver benefits on three levels:to the worker,who gains in personal development and job satisfaction; to the organization, which reforms as staff find new problems to solve and opportunities to pursue; and to the community as a whole, which harvests the economic benefits of hosting positive organizations and workers. This is what makes good jobs productive for the organization, as well as engaging and fulfilling for the worker. Does the ongoing advance of AI threaten to get rid of all the learning, creativity,and meaning that make a job a good job? Certainly, some have blamed technology for just such an outcome. Headlines today often express concern over technological innovation resulting in bad jobs for humans, or even the complete disappearance of certain professions. Some fear that further technology advancement in the workplace will result in jobs where employees are being asked to work in split times or for longer periods over more days. The problem here isn’t the technology; rather, it’s the way the technology is used —and, more than that, the way people think about using it. 1．What is the passage mainly about? A．How to take care of the AI machines. B．The relations between workers and jobs. C．The human’s fears in the age of AI. D．Human experiences were replaced by AI. 2．What jobs are supposed to be good according to the passage? A．Those that mainly have the economic benefits. B．The ones that offer chances to progress professionally. C．Those that make full use of workers’ motivation. D．The ones that don’t bring new problems to workers. 3．How can good jobs bring about benefits? A．By encouraging the workers to learn while doing. B．By providing the organization with pursuing opportunities. C．By solving new problems among the workers. D．By balancing work and job hours in the community. 4．How do news media respond to the development of AI? A．Negative. B．Supportive. C．Optimistic. D．Unknown. 5．What will the author probably talk about next? A．The future jobs humans will do. B．The application of AI in the future. C．The ways that humans do jobs. D．The proper attitude towards AI. 【答案】1．C 2．B 3．A 4．A 5．D 【导语】本文是一篇议论文。文章主要介绍了人类对人工智能发展的前景的担忧以及作者认为我们该持有的态度。 1．主旨大意题。根据文章第一段“When it comes to work, workers, and jobs, much of the concern of the modern age boils down to the fear that we’re witnessing the final stage of the game, and that there will be nowhere for humans to withdraw as machines take over the last few tasks. The most recent example comes from the use of artificial intelligence(AI). Is there no area of human experience that can’t be replaced by AI? And if not, what is left for humans to do except the tasks involved in taking care of the machines?(当谈到工作、工人和工作时，现代人的许多担忧归结为一种恐惧，即我们正在见证这场游戏的最后阶段，人类将无处可退，因为机器接管了最后几项任务。最近的例子来自人工智能(AI)的使用。人类的经验是否没有不能被人工智能取代的领域？如果不是，除了照顾机器的任务，人类还能做什么呢？)”以及全文内容可知，文章主要讲述了人类对人工智能发展的担忧。故选C。 2．细节理解题。根据文章第二段“At the heart of this concern is our desire for good jobs that make the most of workers’ natural abilities and where the work provides the worker with motivation and work-life balance. More importantly, good jobs support workers in learning by doing——and, in so doing, deliver benefits on three levels: to the worker, who gains in personal development and job satisfaction; to the organization, which reforms as staff find new problems to solve and opportunities to pursue; and to the community as a whole, which harvests the economic benefits of hosting positive organizations and workers.(这种担忧的核心是我们对好工作的渴望，这种工作能最大限度地发挥员工的天赋，并为员工提供动力和工作与生活的平衡。更重要的是，好的工作支持员工在实践中学习——这样做可以在三个层面上带来好处：对员工来说，他们获得了个人发展和工作满意度；对组织而言，随着员工找到了需要解决的新问题和追求的新机会，组织也在发生变；对整个社区来说也是如此，它从接纳积极的组织和工人中获得了经济效益。)”可知，那些提供职业发展机会的工作才是好的工作。故选B。 3．细节理解题。根据文章第二段“At the heart of this concern is our desire for good jobs that make the most of workers’ natural abilities and where the work provides the worker with motivation and work-life balance. More importantly, good jobs support workers in learning by doing——and, in so doing, deliver benefits on three levels: to the worker, who gains in personal development and job satisfaction; to the organization, which reforms as staff find new problems to solve and opportunities to pursue; and to the community as a whole, which harvests the economic benefits of hosting positive organizations and workers.(这种担忧的核心是我们对好工作的渴望，这种工作能最大限度地发挥员工的天赋，并为员工提供动力和工作与生活的平衡。更重要的是，好的工作支持员工在实践中学习——这样做可以在三个层面上带来好处：对员工来说，他们获得了个人发展和工作满意度；对组织而言，随着员工找到了需要解决的新问题和追求的新机会，组织也在发生变；对整个社区来说也是如此，它从接纳积极的组织和工人中获得了经济效益。)”可知，好的工作支持员工在实践中学习。故选A。 4．推理判断题。根据文章第三段“Headlines today often express concern over technological innovation resulting in bad jobs for humans, or even the complete disappearance of certain professions. Some fear that further technology advancement in the workplace will result in jobs where employees are being asked to work in split times or for longer periods over more days.(今天的头条新闻经常表达对技术创新导致人类失业，甚至某些职业完全消失的担忧。一些人担心，工作场所的进一步技术进步将导致员工被要求分时间工作，或者工作时间更长、工作天数更多。)”可知，新闻媒体对人工智能的发展抱有担忧的态度。故选A。 5．推理判断题。根据文章最后一段“The problem here isn’t the technology; rather, it’s the way the technology is used —and, more than that, the way people think about using it.(这里的问题不在于技术；更确切地说，是技术的使用方式——更重要的是，是人们对使用技术的看法。)”可推测，作者接下来会谈论人们对待AI的正确态度。故选D。 Passage 4 （2023·天津南开·二模）We are encountering real-world examples of how AI can harm human relations. As digital assistants such as Alexa or Siri become popular, we become accustomed to talking to them as though they were alive. Writing in these pages several years ago, Judith Shulevitz described how some of us are starting to treat them as friends and therapists. Shulevitz herself says she confesses (忏悔) things to Google Assistant that she wouldn’t tell her husband. If we grow more comfortable talking to our devices about our secrets, what happens to our human marriages and friendships? Designers and programmers typically create devices whose responses make us feel better — but may not help us be self-reflective or think over painful truths. As AI goes deeper into our lives, we must face the possibility that it will prevent our emotions and deep human connects. Besides, we will fight with some other challenges. The age of driverless cars, after all, is upon us. These vehicles promise to considerably reduce the exhaustion and distraction that put human drivers in danger, thus preventing accidents. But what other effects might they have on people? Driving is a very modern kind of social interaction, requiring high levels of cooperation. I worry that driverless cars, by taking away from us an occasion to exercise this ability, could contribute to its decline. Not only will these vehicles be programmed to take over driving duties and hence to remove from humans the power to make moral judgments (for example, about which pedestrian to hit when a crash is unavoidable), they will also affect humans with whom they’ve had no direct contact. For instance, drivers who have steered awhile alongside an autonomous vehicle traveling at a steady, changeless speed might drive less attentively, thus increasing their likelihood of accidents once they’ve moved to a part of the highway occupied only by human drivers. Alternatively, experience may reveal that driving alongside autonomous vehicles travelling in perfect accordance with (按照) traffic laws actually improves human performance. Either way, we should be careful to launch new forms of AI without first taking such unexpected social effects into account. We must apply the same effort that we apply to the hardware and software that make self-driving cars possible to managing AI’s potential effects on those outside the car. After all, we install brake lights on the back of your car not just, or even primarily, for your benefit, but for the sake of the people behind you. 1．What can be inferred about human relationships from the first paragraph? A．AI will lead to distant inter-personal relationships. B．We will feel comfortable speaking to others online. C．AI will enable people to communicate more with others. D．We will be more self-reflective in interaction thanks to AI. 2．In Paragraph 2, the phrase “its decline” refers to the decline in ________. A．drivers’ interaction with the cars B．drivers’ exhaustion and distraction C．our ability to cooperate with others while driving D．our ability to deal with emergencies while driving 3．According to the passage, which of the following statements is true of driverless cars? A．They may be better at making more judgments than human drivers. B．They need to vary their speed to make contact with human drivers. C．They may make human drivers in other cars drive more safely. D．They need to force human drivers to concentrate in the car. 4．Which of the following is the writer most likely to agree with? A．Brake lights on the back of our car are installed mainly to warn us of danger. B．We should figure out how new technology affects people before developing it. C．We can launch new forms of AI without thinking of unexpected social effects. D．More efforts should be made to advance the hardware and software of driverless cars. 5．What’s the author’s purpose of writing this passage? A．To present the challenges brought by AI. B．To explain the reason why AI may harm human relations. C．To put forward how to solve the unexpected effect of AI. D．To discuss the advantages and disadvantages of driverless cars. 【答案】1．A 2．C 3．C 4．D 5．A 【导语】 本文是一篇议论文。作者通过日常生活中Alexa和Siri的例子，提出自己的论点——人工智能会阻止人们之间更深层次的交流。接着开始提出自己对于无人驾驶汽车的看法和担忧，并提出需要进一步提升无人驾驶汽车的软件和硬件，让它们更好、更安全地服务于人们。 1．推理判断题。根据第一段中的“As AI goes deeper into our lives, we must face the possibility that it will prevent our emotions and deep human connects. (随着人工智能深入我们的生活，我们必须面对这样一种可能性，即它将阻止我们的情感和深层次的人际联系)”可推知，作者认为人工智能会导致人与人之间的联系变得越来越肤浅、浅显。故选A。 2．词句猜测题。根据划线词上文“But what other effects might they have on people? Driving is a very modern kind of social interaction, requiring high levels of cooperation.(但它们还会对人类产生什么影响呢？开车是一种非常现代的社会互动，需要高度的合作)”及划线词所在句“I worry that driverless cars, by taking away from us an occasion to exercise this ability, could contribute to its decline.(我担心，无人驾驶汽车夺走了我们锻炼这种能力的机会，可能会导致这种它的衰落)”可知，“its decline”指的是前文“this ability”的衰弱和下降，而“this ability”指代上文中的“high levels of cooperation”，即开车时的合作能力。故选C。 3．细节理解题。根据第三段中的“For instance, drivers who have steered awhile alongside an autonomous vehicle traveling at a steady, changeless speed might drive less attentively, thus increasing their likelihood of accidents once they’ve moved to a part of the highway occupied only by human drivers. Alternatively, experience may reveal that driving alongside autonomous vehicles travelling in perfect accordance with (按照) traffic laws actually improves human performance.(例如，在一辆以稳定不变速度行驶的自动驾驶汽车旁边驾驶了一段时间的司机可能会不太注意驾驶，因此，一旦他们行驶到只有人类司机占据的高速公路路段，发生事故的可能性就会增加。经验表明，与完全遵守交通法规行驶的自动驾驶汽车并排行驶实际上会提高人的表现)”可知，无人驾驶汽车能降低其他车辆中的司机出现交通事故的概率，也就是说它们能使人类驾驶员在其他车辆中更安全。故选C。 4．推理判断题。根据最后一段中的“We must apply the same effort that we apply to the hardware and software that make self-driving cars possible to managing AI’s potential effects on those outside the car. After all, we install brake lights on the back of your car not just, or even primarily, for your benefit, but for the sake of the people behind you.(我们必须付出使自动驾驶汽车成为可能的硬件和软件所付出的同样努力，来管理人工智能对车外人员的潜在影响。毕竟，我们在你的车后面安装刹车灯不仅仅是，甚至主要是为了你的利益，而是为了你后面的人)”可推知，作者认为对于自动驾驶汽车，我们应该努力提高它的硬件和软件，让它更加安全。故选D。 5．推理判断题。根据第一段中的“We are encountering real-world examples of how AI can harm human relations.(我们正在遇到人工智能如何损害人际关系的现实例子)”和第二段中的“Besides, we will fight with some other challenges. The age of driverless cars, after all, is upon us.(此外，我们将与其他一些挑战作斗争。毕竟，无人驾驶汽车的时代已经来临)”及第三段中的“Not only will these vehicles be programmed to take over driving duties and hence to remove from humans the power to make moral judgments (for example, about which pedestrian to hit when a crash is unavoidable), they will also affect humans with whom they’ve had no direct contact.(这些车辆不仅会被编程为接管驾驶职责，从而剥夺人类做出道德判断的权力(例如，当碰撞不可避免时，该撞哪位行人)，它们还会影响到与它们没有直接接触的人类)”可推知，作者写这篇文章的目的是展示人工智能带来的挑战。故选A。 【同源语篇阅读】 【材料来源：The New York Times February 6, 2025】 Why DeepSeek Could Change What Silicon Valley Believes About AI The artificial intelligence breakthrough that is sending shock waves through stock markets, spooking Silicon Valley giants, and generating breathless takes about the end of America's technological dominance arrived with an unassuming, wonky title: "Incentivizing Reasoning Capability in LLMs via Reinforcement Learning." The 22-page paper, released last week by a scrappy Chinese AI start-up called DeepSeek, didn't immediately set off alarm bells. It took a few days for researchers to digest the paper's claims, and the implications of what it described. The company had created a new AI model called DeepSeek-R1, built by a team of researchers who claimed to have used a modest number of second-rate AI chips to match the performance of leading American AI Models at a fraction of the cost. DeepSeek said it had done this by using clever engineering to substitute for raw computing horsepower. And it had done it in China, a country many experts thought was in a distant second place in the global AI race. Some industry watchers initially reacted to DeepSeek's breakthrough with disbelief. Surely, they thought, DeepSeek had cheated to achieve R1's results, or fudged their numbers to make their model look more impressive than it was. Maybe R1 was actually just a clever re-skinning of American AI models that didn't represent much in the way of real progress. Eventually, as more people dug into the details of DeepSeek-R1 - which, unlike most leading AI models, was released as open-source software, allowing outsiders to examine its inner workings more closely - their skepticism morphed into worry. And late last week, when lots of Americans started to use DeepSeek's models for themselves, and the DeepSeek mobile app hit the number one spot on Apple's App Store, it tipped into full-blown panic. Based on conversations I've had with industry insiders, and a week's worth of experts poking around and testing the paper's findings for themselves, it appears to be throwing into question several major assumptions the American tech industry has been making. The first is the assumption that in order to build cutting-edge AI models, you need to spend huge amounts of money on powerful chips and data centers. It's hard to overstate how foundational this dogma has become. Companies like Microsoft, Meta and Google have already spent tens of billions of dollars building out the infrastructure they thought was needed to build and run next-generation AI models. They plan to spend tens of billions more - or, in the case of OpenAI, as much as $500 billion through a joint venture with Oracle and SoftBank that was announced last week. DeepSeek appears to have spent a small fraction of that building R1. The obvious conclusion to draw is not that American tech giants are wasting their money. It's still expensive to run powerful AI models once they're trained, and there are reasons to think that spending hundreds of billions of dollars will still make sense for companies like OpenAI and Google, which can afford to pay dearly to stay at the head of the pack. 【译文欣赏】 为什么DeepSeek可能改变硅谷关于AI的认知 一项人工智能突破在股市掀起冲击波、令硅谷巨头感到恐慌、引发了关于美国技术主导地位终结的热议，这项突破却以一个低调、学究气的标题出现：“通过强化学习激励大语言模型的推理能力”。 这份22页的论文由一家结构松散、名为DeepSeek的中国AI初创公司在上周发布，当时论文并没有立即敲响警钟。 研究人员花了几天时间来消化论文的主张，以及所描述内容的可能影响。 该公司创造了一个名为DeepSeek-R1的AI新模型，构建模型的研究团队声称，他们用数量不多的二流AI芯片、以极低的成本就达到了堪与美国一流AI公司相媲美的性能。 DeepSeek表示，它是通过用巧妙的工程技术来替代原始算力而做到的。 而且它是在中国做到了这一点，许多专家认为中国在全球AI竞赛中处于远远落后的第二位。 一些行业观察家最初对DeepSeek的突破表示怀疑。 他们认为，DeepSeek为了达到R1的结果肯定作了弊，或者篡改了数据，让模型看起来比实际更厉害。 或许R1实际上只是对美国AI模型进行了巧妙的换皮包装，并不代表真正取得了多少进步。 最终，随着越来越多的人深入研究DeepSeek-R1的细节（与大多数领先的AI模型不同，它是作为开源软件发布的，让外部人员能更仔细地审查其内部工作原理），他们的怀疑变成了担忧。 上周晚些时候，当许多美国人开始亲自使用DeepSeek的模型，当DeepSeek移动端应用程序在苹果应用商店排名第一时，这个模型引发了全面的恐慌。 根据我与业内人士的交谈，以及一周以来专家们的探索和对论文结果的亲自测试，这个模型似乎对美国科技行业一直做出的几个主要假设提出了质疑。 第一个假设是，为了构建最尖端的AI模型，你需要在强大的芯片和数据中心上花费巨额资金。 这个教条的根深蒂固怎么夸大都不为过。 微软、Meta、谷歌之类的公司已经花费了数百亿美元来建造他们认为构建和运行下一代AI模型所需的基础设施。 他们计划还要再投入数百亿美元，或者像OpenAI的情况，上周宣布通过与甲骨文和软银的合资企业，再投入多达5000亿美元。 DeepSeek建造R1似乎只花费了这些金额的极小一部分。 可以得出的明显结论并不是美国科技巨头在浪费金钱。 一旦经过训练，运行强大的AI模型仍然花费不菲，而且有理由认为，对于OpenAI和谷歌这样的公司来说，花费数千亿美元仍然是有道理的，因为这些公司有能力付出高昂的代价来保持行业领先地位。 【词汇积累】 source n. 来源；发源地；原始资料 v. 从...获得 impressive adj. 给人深刻印象的；令人钦佩的 inner adj. 内心的；内部的；里面的 n. 里面 technological adj. 科技的；工艺的 disbelief n. 不相信；怀疑 joint adj. 联合的；共同的；连带的；合资的 n. 关节；接头；接合处；接缝；夜总会 v. 贴合；连接 initially adv. 最初 rate n. 比率；速度；价格；费用；等级 v. 认为；估价；定等级；值得；怒斥；责骂 generation n. 一代（人）；一代人的时间；（产品发展的）代；产生 obvious adj.明显的;显著的;平淡无奇的;自明 panic n. 惊慌；恐慌 v. 使…惊慌；感到恐慌 adj. 恐慌的；惊慌的 distant adj. 疏远的；遥远的；远房的；陌生的；远行的 assumption n. 假定；设想；担任(职责等)；假装 raw adj. 未加工的；生的；原始的；湿冷的 n. 触到某人的痛处 breakthrough n. 突破，突破性进展；关键问题的解决 make sense 讲得通； 有意义； 言之有理 leading adj. 领导的；指导的；主要的；在前的 【词汇拓展】 venture n. 冒险(事业)；风险；投机；企业 v. 冒险；敢于；冒昧地说 infrastructure n. 基础；基础设施 digest n. 摘要 vt. 消化；理解 vi. 消化 breathless adj. 喘不过气来的 horsepower n. 马力(功率单位) dearly adv. 深深地；真挚地；昂贵地 overstate v. 夸大，对…言过其实，夸大的叙述 a fraction of 一小部分, 少许 engineering n. 工程学；工程 dominance n. 支配；控制；统治；优势；【生】优势度，显性 full-blown adj. 盛开的；成熟的；全面的 substitute for 用…代替， （使）代替;替换;抵换 dogma n. 教条；教义；信条 skepticism n. 怀疑论；怀疑态度；怀疑主义 ( 1 )原创精品资源学科网独家享有版权，侵权必究！ 学科网（北京）股份有限公司 $$