热点话题01 深度应用AI 与伦理挑战
一、主题解析
“深度应用AI 与伦理挑战”这一主题在英语课程标准中是关于 “科学技术与社会发展” 的子话题，
聚焦人工智能技术在教育、医疗、社交等领域的渗透应用，以及由此引发的隐私保护、算法公平、学术
诚信等伦理争议。这一主题既呼应新课标对 “科技素养与人文关怀” 的核心要求，也是高考议论文阅
读的高频考点（占科技类话题考查比重的 30% 以上），主要考查考生对技术利弊的辩证分析能力、伦理
争议的逻辑梳理能力及跨文化语境下的文本解读能力。
二、句子积累
1. 阅读必备词块
（1）基础技术
artificial intelligence (AI) 人工智能
AI-powered systems 人工智能驱动的系统
machine learning 机器学习
generative AI 生成式人工智能
algorithm 算法
data training 数据训练
neural network 神经网络
AI literacy 人工智能素养
intelligent assistant 智能助手
automated decision-making 自动化决策
（2）伦理核心
ethical principles 伦理原则
do no harm 无害原则
non-discrimination 非歧视性
transparency 透明度
explainability 可解释性
proportionality 比例适度原则
human agency 人类主体性
data privacy 数据隐私
information security 信息安全
ethical dilemma 伦理困境
（3）应用场景
personalized learning 个性化学习
academic integrity 学术诚信
content detection 内容检测
recommendation algorithm 推荐算法
medical diagnosis 医疗诊断
social media algorithms 社交媒体算法
employment displacement 就业替代
inclusive technology 包容性技术
cultural diversity 文化多样性
话题资源
environmental sustainability 环境可持续性
（4）争议问题
algorithmic bias 算法偏见
information cocoon 信息茧房
AI hallucination 算法幻觉
plagiarism risk 抄袭风险
technology dependence 技术依赖
human-machine collaboration 人机协作
regulatory framework 监管框架
ethical controversy 伦理争议
accountability 问责制
technological ethics 科技伦理
（5）高频短语
double-edged sword 双刃剑
ethical guidelines 伦理准则
responsible innovation 负责任创新
misuse of technology 技术滥用
safeguard rights 保障权利
mitigate risks 降低风险
balance efficiency and fairness 平衡效率与公平
preserve authenticity 维护真实性
pose threats to 对…… 构成威胁
adhere to norms 遵守规范
2. 常见句式
（1）现象描述
With the rapid advancement of generative AI, the boundary between human-created and machine-generated
content has become increasingly blurred.（随着生成式人工智能的快速发展，人类创作与机器生成内容的界
限日益模糊。）
AI-powered systems have revolutionized sectors like education and healthcare by enhancing efficiency and
precision.（人工智能驱动的系统通过提升效率与精准度，彻底改变了教育和医疗等领域。）
Social media algorithms are designed to maximize user engagement, often leading to the formation of information
cocoons.（社交媒体算法旨在最大化用户参与度，往往导致 “信息茧房” 的形成。）
The pervasive use of AI in academic writing has sparked widespread debate about the future of academic integrity.
（人工智能在学术写作中的广泛应用，引发了关于学术诚信未来的广泛争论。）
Machine learning models rely on massive datasets, which may inadvertently embed existing social biases.（机器
学习模型依赖海量数据集，这可能无意中嵌入已有的社会偏见。）
（2）利弊分析句
On one hand, AI facilitates personalized learning by adapting to individual students' needs; on the other hand, it
risks reducing face-to-face teacher-student interaction.（一方面，人工智能通过适应学生个体需求促进个性化
学习；另一方面，它可能减少师生面对面互动。）
While AI improves the efficiency of medical diagnosis, it cannot replace the empathy and clinical judgment of
human doctors.（尽管人工智能提升了医疗诊断效率，但它无法替代人类医生的共情能力和临床判断
力。）
AI brings unprecedented convenience to daily life, yet its potential to displace repetitive jobs cannot be ignored.
（人工智能为日常生活带来前所未有的便利，但它替代重复性工作的潜力不容忽视。）
The advantage of AI lies in its ability to process complex data quickly, but its disadvantage is the lack of
transparency in decision-making.（人工智能的优势在于快速处理复杂数据的能力，但其劣势是决策过程缺
乏透明度。）
AI-driven tools enhance research productivity, but their misuse may undermine the sanctity of original academic
work.（人工智能驱动工具提升了研究效率，但滥用可能损害原创学术成果的神圣性。）
（3）伦理争议
One of the major ethical concerns about AI is that algorithmic bias may exacerbate social inequality.（人工智能
的主要伦理担忧之一是算法偏见可能加剧社会不平等。）
The question of whether highly anthropomorphic robots should be granted legal rights remains controversial.（高
度拟人化机器人是否应被赋予法律权利的问题仍存在争议。）
Protecting personal privacy is the basic premise for AI to gain social trust, according to industry experts.（业内专
家表示，保护个人隐私是人工智能获得社会信任的基本前提。）
Ethical norms are not shackles on innovation but guarantees for the sustainable development of AI.（伦理规范不
是创新的枷锁，而是人工智能可持续发展的保障。）
The core ethical dilemma of AI lies in balancing technological progress with the protection of human rights.（人
工智能的核心伦理困境在于平衡技术进步与人权保护。）
（4）解决方案
To address AI ethics issues, governments should establish clear regulatory frameworks and accountability
mechanisms.（为解决人工智能伦理问题，政府应建立明确的监管框架和问责机制。）
Universities have integrated AI literacy into curricula to foster students' critical thinking about technology ethics.
（大学已将人工智能素养纳入课程，以培养学生对科技伦理的批判性思维。）
Tech companies are required to conduct safety assessments throughout the AI model's training and deployment.
（科技公司需在人工智能模型的训练和部署全流程进行安全评估。）
Promoting inclusivity in AI design can mitigate risks to marginalized groups and people with disabilities.（在人
工智能设计中促进包容性，可降低对边缘化群体和残障人士的风险。）
International collaboration is essential to developing universal ethical principles for AI development.（国际合作
对制定人工智能发展的通用伦理原则至关重要。）
（5）观点表达
As experts point out, the ultimate goal of AI development is to benefit all human beings rather than replace them.
（正如专家所言，人工智能发展的终极目标是造福全人类而非替代人类。）
It is widely believed that algorithmic fairness is more important than technological advancement in AI
applications.（人们普遍认为，在人工智能应用中，算法公平比技术先进更重要。）
From the perspective of education, AI should serve as an auxiliary tool rather than a substitute for teachers.（从教
育角度看，人工智能应作为辅助工具而非教师的替代品。）
There is no consensus on how to define the ethical boundaries of generative AI among scholars.（学者们对如何
界定生成式人工智能的伦理边界尚未达成共识。）
In terms of data security, strict measures must be taken to prevent the misuse of user information by AI systems.
（在数据安全方面，必须采取严格措施防止人工智能系统滥用用户信息。）
（6）趋势预测
The future of AI will witness closer human-machine collaboration rather than one-sided replacement.（人工智能
的未来将见证更紧密的人机协作，而非单方面替代。）
With the improvement of ethical regulations, AI is expected to develop in a more responsible and sustainable way.
（随着伦理法规的完善，人工智能有望以更负责任、更可持续的方式发展。）
It is predicted that AI will reshape industry patterns but cannot replace human creativity and imagination.（据预
测，人工智能将重塑行业格局，但无法替代人类的创造力和想象力。）
The integration of AI with ethical education will become a key trend in future school curricula.（人工智能与伦
理教育的融合将成为未来学校课程的重要趋势。）
As AI becomes more pervasive, public awareness of its ethical risks will continue to grow.（随着人工智能日益
普及，公众对其伦理风险的认知将持续提升。）
1
Embodied intelligence (具身智能), a revolutionary concept in artificial intelligence (AI), challenges the
traditional view that intelligence exists purely in algorithms (算法). Unlike conventional AI systems that process
data in isolation, embodied AI interacts with the physical world through sensors and actuators, imitating how
humans learn by integrating body and mind.
Research from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) reveals that robots
with embodied intelligence outperform screen-bound AI in complex tasks. For example, a robot arm learning to
突破演练
pour water adjusts its movements based on tactile feedback — a process requiring continuous adaptation, which
demonstrates how physical interaction enhances cognitive (认知) development in ways keyboard-based training
cannot achieve. This “learning-by-doing” approach agrees with developmental psychology theories emphasizing
the role of sensory experiences in human intelligence.
However, critics argue that embodied AI faces unique challenges. While digital AI can be trained on
massive datasets, physical robots require costly real-world trials. A 2023 study in Nature Robotics found that an
embodied AI system took 40% longer to master object recognition than its virtual counterpart, partly because
environmental variables like lighting and friction (摩擦) introduced unpredictability. Hardware durability further
complicates application; saltwater exposure reduces marine exploration robots’ lifespan by 60% compared to lab
models. Additionally, ethical concerns emerge as these systems gain autonomy — should a self-driving car
prioritize passenger safety or pedestrian lives in unavoidable crashes?
Despite these disadvantages, industries are embracing embodied intelligence. Medical robots now assist
surgeons by providing tactile (触觉的) feedback during operations, while agricultural drones analyze soil quality
through aerial imaging and physical sampling. As Dr. Cynthia Breazeal, a pioneer in social robotics, observes:
“True intelligence isn’t just about dealing with numbers; it’s about understanding and navigating the messy reality
we live in.”
The future of AI may depend on bridging the gap between digital and physical gaps. Emerging technologies
like neuromorphic computing, which imitates the brain’s neural structure, could enhance embodied systems’ real-
time adaptability. For tasks demanding contextual awareness and adaptive problem-solving, embodied intelligence
offers solutions that no screen can reproduce.
1．What can we know about embodied intelligence?
A．Physical interaction enhances its cognitive efficiency.
B．Screen-bound AI outperforms it in adaptability.
C．It prioritizes algorithmic complexity.
D．It processes data in isolation.
2．What challenge does the 2023 study in Nature Robotics highlight for embodied intelligence?
A．Timely tactile feedback.
B．Limited dataset accessibility.
C．Unpredictable environmental factors.
D．Too much reliance on virtual training.
3．Why is Cynthia Breazeal mentioned in the passage?
A．To stress ethical concerns in AI.
B．To demonstrate the popularity of robots.
C．To clarify hardware durability challenges.
D．To prove the feature of embodied intelligence.
4．What is the attitude of the author to embodied intelligence?
A．Indifferent.
B．Objective.
C．Dismissive.
D．Skeptical.
2
A company backed by Sam Altman, Tools for Humanity, has introduced a groundbreaking solution to a
pressing problem on the Internet: distinguishing humans from AI bots. The solution is called the Orb, a facial
scanning device designed to verify human identity in a world where AI can easily bypass traditional CAPTCHA
systems.
The Orb is part of the “World” network, an open-source platform for verified humans. When users scan
their faces with the Orb, it analyzes their eyes and facial dimensions, as well as how their features react to light.
Once verified, the Orb sends a unique “WorldID” to the user’s phone, which can serve as a digital passport for
various online services. This technology aims to create bot-free online spaces, preventing fake accounts and
reducing the spread of disinformation.
Currently, the Orb is available in over 20 countries, with more than 12 million people having verified their
identities using the device. It is temporarily unavailable in the US while the technology is updated. The company
plans to roll out home delivery of Orbs later this year, starting in Latin America.
However, the Orb has faced significant challenges. Privacy concerns have led to a ban in Hong Kong,
China. And data regulators in the UK have launched inquiries into the device. Critics argue that the Orb’s use of
biometric data could pose risks to users’ privacy. While the project has drawn skepticism over its potential to
centralize power, Tools for Humanity claims that the technology is designed to protect users’ privacy. Photos taken
by the Orb are deleted from the device after being sent to the user’s phone, and the data is encrypted (加密) during
transmission.
As the AI era progresses, the Orb represents a bold attempt to balance security and privacy in the digital
world. Tiago Sada, the company’s chief product officer, believes the need for such a service is urgent. He points
out that bots are rampant in ticket sales, dating apps, and social media, driving up prices and spreading
misinformation. Whether it can overcome regulatory hurdles and gain widespread acceptance remains to be seen.
1．Which of the following is a function of the Orb?
A．It can be used to bypass traditional CAPTCHA systems.
B．It generates a special ID for users after verifying their identity.
C．It allows users to access restricted areas in the US.
D．It helps users verify their identity for different types of services.
2．What is the main purpose of introducing the Orb?
A．To solve the problems caused by bots online.
B．To create a digital passport for online services.
C．To centralize power through advanced technology.
D．To promote the use of facial scanning devices globally.
3．What can be inferred about the Orb’s impact on privacy?
A．The Orb’s data encryption during transmission is highly effective.
B．The Orb’s biometric scanning may increase privacy risks.
C．The Orb’s been banned in all regions due to privacy concerns.
D．The Orb’s developers are indifferent to privacy issues.
4．What does Tiago Sada’s statement imply about the current online environment?
A．It is significantly impacted by the presence of bots.
B．It is increasingly difficult to distinguish humans from bots.
C．It is facing growing concerns over privacy issues.
D．It is struggling to adapt to the rapid development of AI technology.
3
My struggle with electromagnetic principles nearly ended my physics journey last term. Midterm
preparations found me bewildered by textbook illustrations of Fleming’s Left-Hand Rule, my notebook littered
with frustrated erasures, and my test scores stayed very low. Everything changed when my teacher introduced
DeepSeek, an AI study assistant designed for high school students.
The first time I used it, I typed: “Why does moving a wire near a magnet produce electricity?” Instead of
complicated explanations, DeepSeek responded with a cartoon video showing a simple generator. Then it asked:
“What if we use a stronger magnet? Will the light bulb glow brighter?” These step-by-step questions made me
think like a scientist.
Over weeks, DeepSeek became my personal coach. It noticed I kept mixing up “voltage” and “current,” so
it created a mini-game comparing electricity to water flow. My favorite feature was the “Real-World Connection”
button---it explained traffic light sensors using the same physics principles I learned.
The results surprised everyone. In the midterm exam, I scored 82%! Mr. Li, my physics teacher, pointed at
my paper: “Your explanation of battery energy loss shows true understanding, not just memorization.” Even my
classmates asked how I improved so fast.
Now, I spend 20 minutes daily on DeepSeek. It won’t give direct answers before tests, but guides me to
textbook pages or short animations. Sometimes it connects math problems too - last week it used pizza slices to
teach fractions (分数)!
This experience taught me two things: First, asking questions is more important than memorizing answers.
Second, technology works best when it supports, not replaces our learning journey. With tools like DeepSeek,
every student can find their own path to success.
1．What initially challenged the author in physics?
A．Time management.
B．Laboratory experiments.
C．Electromagnetic concepts.
D．Mathematieal calculations.
2．How did DeepSeek explain electricity principles in Paragraph 2?
A．Through textbook quotes.
B．By providing exam papers.
C．Showing real generator models.
D．Using visual aids and guiding questions.
3．How did Mr. Li find the author’s exam answer?
A．It was perfectly memorized.
B．It used advanced vocabulary.
C．It showed genuine comprehension.
D．It included precisely varied formulas.
4．What does the underlined part “supports, not replaces” in the last paragraph suggest?
A．Human effort remains essential.
B．Technology is indeed unreliable.
C．AI should take control of all learning.
D．Teachers will completely lose their jobs.
4
Becky Litvintchouk has ADHD, which makes it hard for her to focus and organize tasks. She started a
business called GetDirty, which makes cleaning wipes. To help with her business, she uses an app called Claude,
which is an AI tool. This app helps her choose the best contracts and make business plans.
People with ADHD often struggle with focusing, organizing, and controlling their actions. AI tools can help
them do tasks faster. But experts say these tools should not replace traditional treatments for ADHD, like learning
organizational skills or taking medicine. John Mitchell, a professor at Duke University, says AI should be used as
“one tool in a toolbox” He compares it to a life preserver in water; it helps you stay afloat but doesn’t teach you
how to swim.
Becky also uses ChatGPT for shopping and meal planning, which are tasks that can be hard for people with
ADHD．She shares her methods on TikTok to help others.
Bram de Buyser, a software engineer, created Goblin tools, which helps with making to-do lists. He says
he’s not trying to cure ADHD but to help for a few minutes each day.
However, there are concerns about using AI. Russell Fulmer, a professor at Husson University, says the
research on AI and ADHD is not clear yet. He worries that AI might not work well for everyone.
Valese Jones, who has ADHD, uses AI bots (机器人程序) for emails and proofreading. But she says the
responses don’t always show her true self or her culture.
Using AI chatbots has its benefits, like never being tired or needing sleep. But there are also privacy issues.
When you use these bots, you give a lot of personal information to big companies.
In summary, AI tools can help people with ADHD manage tasks better, but they should be used carefully
and not replace traditional treatments.
1．What can we learn about AI from Paragraph 2?
A．It is the most effective way to treat ADHD．B．It is a replacement for traditional treatments.
C．It is a tool to help finish tasks.
D．It is the underlying solution.
2．What does Bram de Buyser aim to do with Goblin tools?
A．Help ADHD patients daily.
B．Teach people to accept the reality.
C．Cure ADHD．
D．Reduce the pain.
3．Which of the following would Russell Fulmer agree?
A．AI should be used as the only treatment. B．The research on AI and ADHD is conclusive.
C．AI works perfectly for people with ADHD．D．Not everyone with ADHD can benefit from AI.
4．How does Valese Jones feel about her using AI bots?
A．She has a mixed feeling about it.
B．She is always satisfied with it.
C．She is uninterested in it.
D．She is totally disappointed with it.
5
AI is advanced enough to take over the basic function of driving a vehicle, but driving needs so much more
than just piloting a car. Human beings perform a vast variety of tasks in a car.
The fact is, human beings won’t completely be out of the driver’s seat anytime soon. We still have the edge
over AI because we think like other human beings. We anticipate that other drivers or pedestrians may not always
follow the rules of the road, or they may behave in unpredictable ways. We know this not from a driving manual,
but from life experience. We subconsciously respond to situations in a very human way. Eventually, learning
algorithms (算法) will be developed to imitate some of our gut (本能的) reactions, but this will take time.
So far, there’s been little effort on the part of either the automobile industry or public safety officials to
bring the public up to speed: When people hear the term autonomous car,
they are often under the mistaken belief that the role of the human driver has been removed,that he or she is
now downgraded to the role of passenger.
“The biggest misconception about automation is that the more automation there is, the less human
competence is needed. Actually, the more you automate, the more you need to educate, where, when, how, etc.,”
observes Bryan Reimer, a research scientist at MIT’s Center for Transportation and Logistics.
He has studied driver behavior as it relates to automation and raises an interesting point:as our automobiles
assume more and more of the tasks of their operation, there is a risk that human beings will lose some of the real-
world driving experiences that make us competent drivers. This is fine as long as the car remains in control. But in
a world of semi-autonomous vehicles, the drivers may become less equipped to handle the complicated operation
that would stump (难倒) the software.
“The way we approach driver training will likely need an upgrade to accommodate to the new world of
semi-autonomous cars. It doesn’t look like we’ re going to have to shut down our driving schools anytime soon,”
Bryan Reimer added.
1．What does the author say about driving?
A．Its basic function is still complicated for AI.
B．It requires more than operating a vehicle.
C．It involves many repetitive and boring tasks.
D．It needs the driver’s full attention to the road.
2．What does the author think is the advantage of human drivers over AI?
A．They prove to be more acceptable to the general public.
B．They have more detailed knowledge of the rules of the road.
C．They are more experienced in dealing with unusual situations.
D．They are better at handling traffic accidents in a human way.
3．What does Bryan Reimer imply about automation?
A．It plays an important role in improving traffic safety.
B．It reduces the need for drivers’ professional competence.
C．It alters the relationship between drivers and passengers.
D．It may cause drivers to be less able to cope with complex situations.
4．What should people do to adapt to semi-autonomous driving?
A．Upgrade automation technologies.
B．Change their perception of AI.
C．Update driver training programs.
D．Open more driving schools.
6
Whether or not robots can feel is a question. But what’s becoming increasingly clear, says Associate
Professor Szu-chi Huang in Stanford University, is that robots do have the capacity to make humans feel.
In the latest episode of If/Then: Business, Leadership, Society, Huang digs into the effect that robots can
have, not just on our emotions, but on our behavior. Huang’s research shows that when people witness others
being helpful towards others, they’re motivated to do the same. This is what she calls “pro-social” behavior. But
she wondered: what happens when a robot is the one lending a helping hand? Are people inspired to follow the
example of a machine?
To find out, Huang designed a study where participants were shown various news reports about natural
disasters and the measures being taken in response. In some stories, the “heroes” were humans; in others, they
were robots.
“In both cases, we explained in detail what those heroes were doing,” says Huang. Whether dragging
survivors out of ruins after an earthquake or disinfecting (消毒) hospitals during a surging COVID-19 pandemic,
“The actions are exactly the same, but the heroes are different.”
Following test subjects’ exposure to these stories, Huang measured their willingness to engage in pro-social
behavior, like donating to support children in need. What she found was those who saw robot heroes were
significantly less likely to donate than those who saw humans taking the same actions. “The robot stories actually
make people feel less inspired,” says Huang. “And that has important consequences. If using robots lowers our
intention to help others, it could have a pretty big negative social impact.”
So what do we do as AI and robots play an increasing role in our lives? How do we embrace their benefits
without downgrading our humanity and pro-sociability in the process? Huang says if we want robots to be good
for society, then we need to humanize them.
1．Which of the following best describes “pro-social” behavior?
A．Helping others when they are in need.
B．Becoming less motivated to help others.
C．Copying the behavior of a robot.
D．Avoiding social interaction with others.
2．What were the participants asked to do in the study?
A．To review news reports.
B．To respond to news reports.
C．To observe robots working.
D．To take an interview.
3．What can be inferred from the last paragraph?
A．People hate robots helping others.
B．Robots will never be able to replace humans.
C．People’s humanity will surely decline because of robots’ development.
D．Humanizing robots is likely to make them more beneficial to society.
4．Which of the following could be the best title?
A．Robots: Threats to Human Pro-social Behavior
B．The Dilemma of Robots in Modern Society
C．Humanizing Robots: A Solution to Social Problems
D．Do Robots Inspire Human Pro-social Behavior?
7
Superhuman artificial intelligence has played a role in our life. When it comes to playing games like chess,
or solving difficult scientific challenges like predicting protein structures, computers are well ahead of humans.
But we have one superpower they aren’t close to mastering: mind reading.
We are blessed with a mysterious ability to reason the goals, desires and beliefs of others, a crucial skill that
means we can anticipate other people’s actions and the consequences of our own. If AIs are to become truly useful
in everyday life — to cooperate effectively with us or to understand that a child might run into the road after a
bouncing ball — we have to give them this gift that evolution has given us to read other people's minds.
In psychology, the ability to infer another’s mental state is referred to as theory of mind. In humans, this
capacity develops at a very young age. How to reproduce the capability in machines is far from clear, though. One
of the main challenges is context. For instance, if someone asks whether you are going for a run and you reply
“it’s raining”, they can quickly conclude the answer is no. But this requires huge amounts of background
knowledge about running, weather and human preferences.
Besides, whether humans or AI, the theory of mind is supposed to emerge naturally from one’s own
learning process. Building prior knowledge into AI makes it reliant on our imperfect understanding of theory of
mind. In addition, AI may be capable of developing approaches we could never imagine. There can be many
forms of theory of mind that we don’t know about simply because we live in a human body that has certain types
of senses and a certain ability to think.
But we might still want AI to have a more human-like form of theory of mind. Humans can clearly explain
their goals to each other using common language and ideas. While letting AI form the theory of mind in their
learning process is likely to lead to developing more powerful AI, plainly building in shared ways to represent
knowledge may be crucial for humans to trust and communicate with AI.
It is important to remember, though, that the pursuit of machines with theory of mind is about more than
just building more useful robots. It is also a stepping stone on the path towards a deeper goal for AI and robotics
research: building truly self-aware machines. Whether we will ever get there remains to be seen. But along the
way thinking about other people and other agents, we are on the path to learning to think about ourselves.
1．According to the passage,which of the following contexts can AI understand well?
A．When asked where an object is, a naughty kid points in the wrong direction.
B．When a teacher asks for a boy’s homework, he answers “my dog ate it”.
C．When you are treated with a hotpot for dinner, you reply “a sore throat”.
D．When a mom tells her kid milk is good for health, the kid drinks it up.
2．Which statement will the author agree with most about theory of mind?
A．Humans’ theory of mind is far from perfect.
B．Humans limit AI’s theory of mind to an extent.
C．AI’s theory of mind development has been totally decoded.
D．Shared forms of theory of mind result in more powerful AI.
3．What is the author’s attitude toward truly self-aware machines?
A．Ambiguous.
B．Conservative.
C．Skeptical.
D．Approving.
4．Which would be the best title for the passage?
A．AI’s Theory of Mind Will Define Our Future
B．AI with Its Own Theory of Mind Is Expected
C．AI’s Theory of Mind — a Blessing or Suffering to Humans
D．Theory of Mind Bridges the Gap Between Humans and AI
8
Applying for a new job? A human resources representative or hiring manager may not be the first to review
your resume (简历). Instead, it might be an AI-powered tool.
With some job postings attracting hundreds or thousands of applications, companies are increasingly using
AI to sort through applicants before they reach a human recruiter (招聘者). According to data from jobseeker
resource site Resume Genius, 48% of hiring managers report using AI to screen resumes and applications.
Companies have long been using applicant tracking systems (ATS) to analyze resumes for skills and keywords.
Now, AI can further automate this by “reading” resumes to find suitable candidates while helping recruiters save
time.
But can that make it more difficult to guarantee a resume gets seen by a human? Experts claim that AI
enables algorithms (算法) to comprehensively “view” applications. This lessens the dependence on keyword-
focused approaches that might have disadvantaged candidates who didn’t include specific words in their resumes.
The practice of including keywords from a job posting in a resume isn’t new. For example, an applicant
may put the coding languages Python or C + + on their resume for software engineering jobs that require those
skills, ideally improving their chances of being picked up by automated systems. But new AI tools can understand
the content of a resume, meaning they can help spot candidates even if their resumes don’t mention certain words,
said Madeline Laurano, chief analyst at Aptitude Research, an HR advisory firm. The system can then provide
hiring managers with a list of relevant candidates and rank them, added Laurano.
But as advanced as AI may be, AI sometimes outputs false information, known as hallucinating, so it could
potentially misunderstand an applicant’s resume and provide incorrect information to a recruiter. While AI tools
like Hiring Assistant can help remove “friction” in the hiring process, finding the right hire will always require a
human touch.
1．What can be inferred about the AI recruitment?
A．It benefits recruiters and candidates.
B．It focuses upon keyword matching.
C．It helps candidates analyze resumes.
D．It has reduced application numbers.
2．Why do applicants usually add some keywords to their resumes?
A．To highlight personal achievements.
B．To get noticed more likely by ATS.
C．To seek more attention from humans.
D．To let AI tools prioritize reading them.
3．What’s the author’s attitude to AI’s making a decision on hiring?
A．Dismissive.
B．Favorable.
C．Disapproving.
D．Objective.
4．What’s the suitable title for the text?
A．What AI Can Do for Hiring Decisions
B．AI Has Replaced Human Recruiters
C．How AI Transforms Job Applications
D．AI Might Be Reading Your Resume
9
Science chatbots provide a unique way to interact with complex scientific topics, giving users instant
answers, guidance, and. support. Below, you’ll find a list of science-focused chatbots that can help you understand
topics in-depth, conduct research, and even complete experiments.
Wolfram Alpha
Wolfram Alpha is a powerful computational chatbot, perfect for students tackling challenging science and
math problems. Known for its accuracy, Wolfram Alpha provides step-by-step solutions to complex equations,
generates graphs, and interprets data. It’s ideal for students looking to enhance their problem-solving skills in
physics, chemistry, and other quantitative sciences. The platform’s precise calculations and vast database make it a
go-to tool for those needing scientific support.
Socratic
Socratic is a popular educational chatbot developed by Google, helping students with science, math, and
other subjects. It uses AI to answer questions, explain concepts, and provide visual aids for better comprehension.
Socratic is user-friendly and offers solutions for both simple and complex scientific problems. Via images, videos,
and diagrams, Socratic enhances understanding and is a great resource for students who benefit from a visual
learning approach.
Galactica
Galactica is an astronomy-focused chatbot designed to help students and enthusiasts explore the universe. It
provides information on bodies, space phenomena, and astronomical theories, making it an exciting resource for
anyone interested in space science. With Galactica, users can ask questions about planets, stars, and galaxies,
gaining a better understanding of our universe. This chatbot is perfect for aspiring astronomers and students keen
on expanding their knowledge of space.
SciNote ELN Assistant
SciNote ELN Assistant is an electronic lab notebook (ELN) tool tailored for scientific research and lab
management. It assists researchers by organizing notes, tracking experiments, and providing reminders for lab
tasks. This chatbot is particularly beneficial for students and researchers who need to maintain organized records
of their experiments. With SciNote ELN, users can ensure efficient documentation, making it easier to analyze
results and improve research productivity.
1．What is Wolfram Alpha featured with?
A．Interactive science learning visuals.
B．Real-time research database updates.
C．In-depth astronomy exploration guide.
D．Detailed guidance to scientific problems.
2．Which chatbot best suits a visual learner?
A．Wolfram Alpha.
B．Socratic.
C．Galactica.
D．SciNote ELN Assistant.
3．What is the primary function of SciNote ELN Assistant?
A．Analyzing chemical reactions in labs.
B．Generating graphs from research data.
C．Documenting research and tracing lab tests.
D．Providing video explanations of lab techniques.
10
Before she goes to bed each night, Li Juan logs on to DeepSeek for “therapy (疗法) sessions”. Ever since
January, when the breakout Chinese AI app launched, the 28-year-old has brought her dilemmas and sorrows,
including the recent death of her grandmother, to the chatbot as a therapy. Its responses at times brought her to
tears. “DeepSeek has been such an amazing counselor (咨询师). It has helped me look at things from different
perspectives and does a better job than the paid counselling services I have tried," says Li Juan.
From writing reports and Excel formulas to planning trips, workouts and learning new skills, AI apps have
found their way into many people’s lives across the world. In China, though, young people like Li Juan have been
looking to AI for something not typically expected of computing and algorithms — emotional support. For users
like Li Juan, DeepSeek isn’t just a tool; it’s a source of comfort in uncertain times.
DeepSeek is available 24/7 (24 hours a day, 7 days a week), allowing users to seek emotional support
whenever they need it. This is particularly important for those experiencing sudden emotional depression or who
may not have access to traditional therapy during off-hours. And the anonymity (匿名) and privacy offered by the
app make it easier for them to share their thoughts and emotions without hesitation.
DeepSeek is a generative AI tool — similar to OpenAI’s ChatGPT and Google’s Gemini — trained on
massive amounts of information to recognize patterns. This allows it to predict things like people’s shopping
habits, create new content in text and images, and also carry on conversations like a person. What’s more, it lets
users see its “thought process” before delivering a response because it offers something unique: its AI model, R1.
DeepSeek R1 is designed to interpret emotional clues from user interactions, including tone of voice,
language patterns, behavior indicators and contextual understanding. The ability to detect emotions marks a
significant advancement in AI assistants, enabling more empathetic and personalized interactions that adapt to
users’ emotional needs.
1．What does DeepSeek mean to Li Juan?
A．An online teammate.
B．An emotional comfort.
C．An academic guide.
D．A decision maker.
2．What sets DeepSeek apart from other Al apps according to the text?
A．24/7 availability.
B．Massive data training.
C．A wide range of applications.
D．Thinking process visualization.
3．Which aspect of DeepSeek R1 does the last paragraph cover?
A．Its limitations.
B．Its operating environment.
C．Its significance.
D．Its theoretical foundation.
4．What would be the best title for the passage?
A．DeepSeek: The Rise of Chinese AI App B．How Young Chinese Find Therapy in AI
C．How AI Shapes the Consulting Industry D．DeepSeek: The Future of AI Technology
11
In a world where tech companies often compete to create the next productivity tool, a new device is
breaking away from the pack. Meet Friend: a new wearable AI companion that you wear around your neck. The
small device records your every word and interaction and responds accordingly by text. They cost $99 and are
expected to ship out in early 2025.
The device comes at a time when loneliness and social isolation (孤独) are at record highs. Studies indicate
that 61% of young people in the U.S. suffer from serious loneliness, a statistic that reveals a growing mental
health crisis. While AI reshapes all the ways we interact in jobs, healthcare and entertainment — more and more
people report feeling socially isolated, Schiffmann, Meet Friend’s founder, sees it differently: the device isn’t a
replacement for human relationships but rather a way to fill the gaps when human companionship is hard to come
by.
AI companionship is controversial (有争议的). While some people, like Schiffmann, argue that it can help
people feel less isolated, others worry that AI relationships might displace real human connections, and thus make
people feel more lonely. If you are someone who already struggles with human interactions, the thinking goes,
why would you continue to subject yourself to them when you have an AI friend who's fun and uncomplicated,
and doesn’t come with all the baggage of another person?
But so far, the research doesn’t seem to support this fear. Drake-Maples at Stanford University surveyed
more than 1,000 students who used the AI chatbot Replika, and investigated their feelings of loneliness and
perceived social support. Roughly half of these users reported decreased anxiety, and a feeling of social support.
Nearly a quarter of those surveyed said they are more able to handle stress in their current relationship because of
Replika’s advice.
It is still a new field though, and Drake-Maples says there needs to be guidelines. “I strongly believe that
you do need to have ethical guidelines around AI companions pushing people back when appropriate, towards
human relationships,” she says. These could be gentle reminders like “Hey, you should go chat with somebody
about that” or “Go practice this now with a real human.”
1．What is Meet Friend aimed at?
A．Replacing human relationships.
B．Boosting the productivity at work.
C．Offering a new approach to connection.
D．Detecting early symptoms of mental illness.
2．What does the underlined part “the thinking” in paragraph 3 refer to?
A．AI helps avoid life’s stress.
B．AI worsens feelings of loneliness.
C．Human relationships are hard to keep.
D．It's easy to communicate with an AI friend.
3．What is Drake-Maples’ attitude towards AI companionship?
A．Cautious.
B．Unclear.
C．Supportive.
D．Disapproving.
4．Which of the following is the best title for the text?
A．Can we trust an AI friend?
B．Will AI change human nature?
C．Can an AI friend make us less lonely?
D．Will AI redefine mental health crisis?
12
John von Neumann, who pioneered modern computer architecture, wrote in 1949, “It would appear that we
have reached the limits of what is possible to achieve with computer technology.” Among the countless forms of
computational limit-breaking that have challenged von Neumann’ s prediction is the social psychologist Frank
Rosenblatt’s 1958 model of a human brain’ s neural network. He called his device, based on the IBM 704
mainframe computer, the “Sensor” and trained it to recognize simple patterns. Sensor eventually led to deep
learning and modern artificial intelligence.
Although Stuart Dreyfus, who is a professor at Berkeley, is impressed by the progress made in AI, he
remains skeptical and concerned about certain AI applications, especially large language models, or LLMs, like
ChatGPT. “Machines don’t have bodies, which is limiting and creates risk.” he notes, “It seems to me that in any
area which involves life-and-death possibilities, AI is dangerous, because it doesn’t know what death means.”
According to the Dreyfus skill acquisition model, an inner shift occurs as human know-how advances
through five stages of development: beginner, advanced beginner, competent, master, and expert. “A crucial
difference between beginners and more competent performers is their level of involvement,” the researchers
explained. “Beginners and advanced beginners feel little responsibility for what they do because they are only
applying the learned rules.” If they fail, they blame the rules. Expert performers, however, feel responsibility for
their decisions because as their know-how becomes deeply rooted in their brains, nervous systems, and muscles,
they learn to operate the rules to achieve their goals. They own the outcome.
That close relationship between intelligent decision-making and responsibility is an essential ingredient for
a well-functioning, civilized society, and some say it’s missing from today’s expert systems. Also missing is the
ability to care, to share concerns, to make commitments, to have and read emotions — all the aspects of human
intelligence that come from having a body and moving through the world.
As AI continues to pour so many aspects of our lives, can we teach future generations of expert systems to
feel responsible for their decisions? Is responsibility or care or commitment or emotion — something that can be
obtained from statistical inferences or drawn from the problematic data used to train AI? Perhaps, but even then
machine intelligence would not equal to human intelligence — it would still be something different, as the
Dreyfus also described nearly four decades ago.
1．What can we conclude from the first paragraph?
A．It highlights the role of IBM 704 in the evolution of AI.
B．It describes the development and impact of AI on modern society.
C．It illustrates the continuous breakthroughs in computational capabilities.
D．It introduces the rough history of computer technology and its limitations.
2．What is the main concern that Professor Dreyfus has about AI applications?
A．AI’s progress in language models is too rapid.
B．AI’s lack of physical bodies limits its understanding.
C．AI is not suitable for applications involving creativity.
D．AI’s understanding of death is too advanced for its own good.
3．What is the central theme of paragraphs 3 and 4?
A．The necessity of having a body for the development of advanced skills.
B．The significance of emotional intelligence from a beginner to an expert.
C．The evolution of responsibility and decision-making with skill levels advancing.
D．The importance of physical movement in the development of human intelligence.
4．What can we learn from the passage?
A．The limitations of AI in comparison to human intelligence.
B．The evolution of AI beyond human imagination and its destruction of society.
C．The skill acquisition stages of learning operational rules in humans and AI.
D．The prediction of responsibility in AI and human intelligence decision-making.
