考前押题06阅读理解之人工智能（学生版）_高考一二三轮复习_英语_2026年-英语高考专项复习_2026年高考英语阅读理解突破策略及押题-G437_资料

考前猜押06 阅读理解—人工智能
 
1.押题指导原则：
2026 年教育部命题指导思想：优化试题呈现方式和素材选取，融入科技前沿动态，浸润人文教育元素，加
强项目式、探究式的真实情境问题设计，更好考查学生关键能力、学科素养和思维品质。
2.押题依据分析：
猜押
考点
3 年真题
词数
话题
考情分析
押题
依据
阅读
理解
科普
说明
文
2026 年浙江首
考阅读理解
C、D 篇
C355
D321
C 关于虚假评论的一
项研究D 追踪动物AI
工具SLEAP 应用于研
究植物根系表型，助
力筛选有益基因，这
也是跨学科合作的成
果
近年来，人工智能的发展突飞
猛进，人工智能涉及到日常生活
的各个方面。人工智能在高考阅
读理解中频繁出现，文章主要考
查学生对每个环节关键内容的准
确理解，引导学生了解科技前
沿、合理使用先进技术以及对人
工智能的深入思考。鲜明地彰显
了命题人旨在融入科技前沿动态
的命题导向。
人工智能
说明文频
繁出现在
高考阅读
理解中。
文章主语
考查学生
的推理判
断能力和
对人工智
能的评价
态度。辅
助考查细
节理解和
主旨大
意。
2024 年新高考
I 卷阅读D 篇
D
：
319
D 篇数字生物多样性
记录存在偏见
2024 年新高考
Ⅱ卷阅读D 篇
D
：
319
D 篇介绍了Catriona 
Campbell 所著AI by 
Design: A Plan for 
Living with Artificial 
Intelligence 一书
2023 年新高考
I 卷阅读C 篇
C
：
318
C 篇倡导简单的数字
生活方式
押题预测目录
一、人工智能与哲学
二、人工智能的实际运用
三、对人工智能的态度
四、人工智能与职业
五、人工智能与语言学习
六、人工智能的缺陷
七、规范人工智能的发展
八、人工智能与医疗发展
九、人工智能与农业发展
十、人工智能与教育
 一、人工智能与哲学
押题1
词数
话题
体裁
考点
押题理由
345
人工智能(AI)
与哲学之间
的关系
议论文
细节理解
推理判断
主旨大意
人工智能在高考阅读理解中频繁出现，文章主要考查学生对
每个环节关键内容的准确理解，引导学生了解科技前沿、合
理使用先进技术以及对人工智能的深入思考。鲜明地彰显了
命题人旨在融入科技前沿动态的命题导向。
New scientific understanding and engineering techniques have always impressed and frightened us, and will 
continue to do so. Open AI recently announced that it anticipates superintelligent AI going beyond human abilities 
within this decade. It is accordingly building a new team, and devoting 20% of its computing resources to ensuring 
that the behavior of such AI systems will be aligned with (与……一致) human values.
Beyond engineers and researchers, can philosophers shape AI too? Philosophy has been instrumental to AI 
from its early days. In 1956, the Logic Theorist, the first operational AI program, was created to prove 
mathematical theorems (定理), drawing on ideas from philosophers Alfred North Whitehead and Bertrand Russell. 
Besides, German philosopher Gottlob Frege also laid the groundwork for modern logic and computing, which 
helped create early AI systems.
Today’s AI, based on deep learning, is associated with processing vast quantities of data. Philosophy, 
however, remains relevant. Take Large Language Models (LLMs), like ChatGPT, which generates conversational 
text. At their heart, they track statistical patterns of language use. Something very much like this idea was stated by 
the Austrian philosopher Ludwig Wittgenstein: the meaning of a word is found in its use in the language.
Philosophy also raises big questions about AI’s future. Can AI truly understand language or become 
conscious (有意识的)? Science has so far been unable to fully explain how consciousness arises. Similarly, can AI 
be creative? British scientist Margaret Boden believes AI can produce new ideas but struggles to evaluate them like 
humans do.
As AI grows more powerful, aligning AI with human values is not just a technical challenge but a social one. 
It requires input from philosophers, social scientists, policymakers, and citizen users. British author Jamie Susskind 
even suggests building a “digital republic” — one that rejects the political and economic systems that excessively 
empower tech giants.
AI is also changing philosophy itself. Computational models simulate real-world situations, aiding 
philosophers in tackling social questions. As AI develops, it provides new ways to explore age-old questions, 
ensuring philosophy remains essential in guiding our technological future.
1．What is OpenAI mainly concerned about?
A．The waste of its computing resources.
B．The potential risks of superintelligent AI.
C．The decline of human competitive abilities.
D．The high cost of super intelligence research.
2．What is the main focus of Paragraph 2?
A．The technical breakthroughs in early AI.
B．The philosophy’s historical contributions to AI.
C．The role of mathematical theorems in AI creation.
D．The collaboration between philosophers and engineers.
3．Why does the author mention Wittgenstein’s viewpoint?
A．To explain the linguistic logic of LLMs.
B．To argue philosophers’ superiority in AI research.
C．To illustrate philosophy’s relevance to modern AI.
D．To prove LLMs lack genuine language understanding.
4．Which word best defines the relationship between AI and philosophy?
A．Interdependent.B．Irrelevant.
C．Competitive.
D．One-sided.
二、人工智能的实际运用 
押题1
词数
话题
体裁
考点
押题理由
313
美术领域的 
AI 程序 
Aaron 能自主
创作
说明文
细节理解
猜测词义
人工智能在高考阅读理解中频繁出现，文章主要考查学生对
每个环节关键内容的准确理解，引导学生了解科技前沿、合
理使用先进技术以及对人工智能的深入思考。鲜明地彰显了
命题人旨在融入科技前沿动态的命题导向。
By the late 1960s, abstract painter Harold Cohen had represented Britain at important festivals with his oil 
paintings and was seeking a new challenge. “Maybe there are more interesting things going on outside my studio 
than inside it,” he thought. Cohen turned from the canvas to the screen, using computers to find new ways of 
creating art. In the late 1960s, he created a program that he named Aaron. It was the first artificial intelligence 
software in the world of fine art, and Cohen first presented Aaron in 1974 at the University of California, Berkeley.
Now, with AI dominating the headlines, a new exhibition at New York’s Whitney Museum of American Art, 
“Harold Cohen: Aaron”, running through June 2024, draws attention to Cohen’s pioneering work. The Whitney is 
offering something deeper than most previous Aaron exhibits: a real-time experience, in which visitors can watch 
the software produce art.
Cohen seeded Aaron with all sorts of knowledge: about basic objects, physics, and fundamental techniques 
of drawing. Aaron uses this knowledge to follow instructions, complete tasks, and make decisions like human 
beings — a very different approach from today’s generative AI art programs, which don’t draw from scratch (从头
开始) but rather rely on databases of images. Versions of Aaron still generate output, but anything done after 
Cohen’s death in 2016 is not considered genuine.
The Whitney is performing two versions of Cohen’s software, along with the art that each produced before 
Cohen died. The 2001 version, Aaron KCAT, generates images of figures and plants and projects them onto a wall 
more than ten feet high, while the 2007 version produces jungle-like scenes. The software will also create art 
physically, on paper, for the first time since the 1990s. “It is absolutely thrilling,” said Christiane Paul, the 
museum’s director of digital art, “to have one of those remarkable treasures of digital art in the collection.”
1．What was Aaron born out of?
A．The university’s support.
B．The museum’s donation.
C．Cohen’s imagination.
D．Cohen’s curiosity.
2．What can visitors do in the “Harold Cohen: Aaron” exhibition?
A．Observe Aaron creating art on the spot. B．Generate images with the help of Aaron.
C．Learn about the physical rules in art.
D．See the original 196Os version of Aaron.
3．What makes Aaron different from modern AI art programs?
A．Its ability to assign tasks.
B．Its use of modern techniques.
C．Its capability to make choices.
D．Its dependence on existing database.
4．What does the underlined word “projects” in the last paragraph mean?
A．Plans.
B．Throws.
C．Displays.
D．Draws.
押题2
词数
话题
体裁
考点
押题理由
355
人工智能驱
动的森林恢
复系统
说明文
细节理解
推理判断
主旨大意
人工智能在高考阅读理解中频繁出现，文章主要考查学生对
每个环节关键内容的准确理解，引导学生了解科技前沿、合
理使用先进技术以及对人工智能的深入思考。鲜明地彰显了
命题人旨在融入科技前沿动态的命题导向。
We often imagine the future of AI in grand terms. Yet, some of its most transformative work is happening not 
with flashy robots, but with silent algorithms (算法), working to restore our dying forests. In the Amazon, vast 
areas ruined by deforestation have struggled to regrow. The changed microclimate creates a cruel environment 
where only tough species thrive (茁壮成长), often forming a low-diversity “green desert”. Traditional reforestation 
is slow and has a mixed success rate. 
Enter the “algorithmic gardener”. A team of scientists has developed a drone-based AI system that is 
changing the game. First, drones equipped with high-resolution cameras fly over a degraded area, mapping it in 
detail. They analyze soil composition and remaining vegetation plots. This data is fed into a machine learning 
model trained on healthy forest ecosystems. 
The AI doesn’t just suggest what to plant; it designs how to plant. It generates a precise planting blueprint. 
Instead of orderly rows of a single tree species, the algorithm plans a seemingly chaotic mosaic. It might gather 
together drought-resistant plants on a dry area. It calculates the best timing and spacing between native species, 
copying the complex structure of a natural forest. 
The magic truly unfolds at ground level. Drones don’t merely map; they also act as precision planters. 
Carrying cans of seed pods, they navigate the complex planting blueprint. Each biodegradable pod contains a 
careful mix: the primary tree seed and companion plants. This “precision placement” is crucial. 
But the system’s intelligence extends beyond the initial design. It possesses an adaptive learning capability: 
drone flights monitor the young forest’s progress. Advanced image analysis detects early signs of stress — unusual 
leaf discoloration. The AI can then cross-reference this data with real-time weather information, satellite imagery, 
and its original models. If necessary, it generates a revised plan. 
The AI-designed forests show 50% higher survival rates after two years compared to conventionally planted 
plots. Human beings are using AI as a tool to translate the complex language of nature into actionable plans. The 
“algorithmic gardener” works unseen, but its potential legacy (遗产) is a planet that can breathe a little easier.
1．Why is the Amazon mentioned in paragraph 1?
A．To show its present terrible state.
B．To reveal factors for green desert.
C．To stress the traditional planting method.
D．To provide the background for AI’s application.
2．What does the underlined word “mosaic” probably mean in paragraph 3?
A．Regular pattern.
B．Empty field.
C．Complex mixture.
D．Colorful grassland.
3．What can be learned about the AI system’s adaptive learning capability?
A．Referring to satellite imagery.
B．Tracking real-time weather information.
C．Basing a conclusion on the initial design.
D．Achieving improvement through ongoing feedback.
4．Which of the following is the best title for the text?
A．How AI Is Helping Make a Blueprint
B．When Forests Learn to Breathe with Data
C．How Drones Hold the Future of Photography
D．When the End of Traditional Farming Comes
押题3
词数
话题
体裁
考点
押题理由
354
一种新型人工神经元，其电
压与能耗均与生物神经元相
当，实现了更自然、高效的
“细胞间”通信
说明文
细节理解
推理判断
猜测词义
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
Engineers at the University of Massachusetts have developed an artificial neuron (神经元) like no other. In 
experiments, their newly created model was able to directly communicate with a biological neuron in a remarkably 
lifelike, quiet way. The cell-to-cell flow of electrical information involves the same voltage ( 电压) and energy 
consumption as natural neuron communication. The breakthrough is a first in the field, somewhat reducing the 
distance between our brains and our computers.
Scientists have been developing artificial brain cells and connecting them to complex machines for years 
now, but their simple functions pale in comparison to those in our brains. This newest creation is closer than ever to 
imitating the real thing. “Previous versions of artificial neurons used10 times more voltage — and 100 times more 
power — than the one we have created,” says engineer Jun Yao.
That’s like an artificial neuron “screaming” at the top of its lungs to a typical human neuron. All that 
hollering takes a lot of energy, and if it’s too loud, the receiver may be overwhelmed and miss the key of the 
message.
What Yao has created is an artificial neuron that essentially can “whisper” to a natural one. “Ours registers 
only 0.1 volts, which is about the same as the neurons in our bodies,” says Yao. Other scientists have gotten 
artificial neurons to communicate with real brain cells in less natural ways — for example, through the use of light. 
But this new breakthrough copies a more conventional system.
The key in this case was to use protein nanowires (纳米线) grown by bacteria. Because these structures are 
naturally obtained, they can survive in wet environments like real neurons.
Models like these advance the potential for neuromorphic integration. We currently have all kinds of 
wearable electronic sensing systems, but they are comparatively awkward and inefficient. Every time they sense a 
signal from our body, they have to electrically strengthen it so that a computer can analyze it. That intermediate step 
of strengthening increases both power consumption and the circuit’s complexity, but sensors built with the low-
voltage neurons could do without any strengthening at all.
1．How does the new artificial neuron compare with natural ones?
A．It’s less likely to collapse.
B．It’s more complex.
C．It consumes more energy.
D．It shows equal efficiency.
2．What does the underlined word “hollering” in paragraph 3 most likely mean?
A．Whispering.
B．Shouting.
C．Speeding.
D．Waiting.
3．Which is a crucial factor in the breakthrough?
A．The use of protein nanowires grown by bacteria.
B．The development of new computer programming methods.
C．The discovery of how to use light for neuron communication.
D．The invention of special strengthening equipment for signals.
4．What future application is suggested for the new neurons?
A．Faster computer games.
B．Smaller smartphones.
C．Better wearable devices.
D．Brighter screens.
押题4
词数
话题
体裁
考点
押题理由
360
AI 智能眼镜Horizon X
说明文
细节理解
推理判断
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
When Justin Cheung lost his sight in a car accident at 28, he feared his hiking passion would fade. “Trails 
were my happy place, but I couldn’t even navigate sidewalks safely,” he recalled. Three years ago, he tested 
Horizon X, an AI-powered smart glass by Dr. Elara Voss’s team. Today, Cheung has hiked 500+ miles across the 
Rockies, guided by audio cues like “small rock 3 feet left”  or “gentle uphill 50 yards”.
Horizon X isn’t just a tool — it’s the product of Dr. Voss’s decade-long work. Her grandmother, who lost 
vision to glaucoma in her 70s, inspired her to create a device that “restores independence, not just compensates for 
loss”. Unlike traditional canes or magnifiers, the glasses use advanced computer vision to analyze surroundings in 
real time: detect obstacles, recognize trail markers, and alert users to terrain changes affecting balance. Data is 
translated into clear audio prompts via bone-conduction headphones, letting users track natural sounds key to 
outdoor lovers.
“Early prototypes were bulky and slow — mixing up squirrels and boulders, taking 2 seconds to respond,” 
Dr. Voss laughed. “We partnered with parks to collect 1.2 million trail images, training AI on diverse outdoor 
conditions. Now it distinguishes twigs from cliff edges in 0.3 seconds.” For Cheung, it changed his life. On his first 
hike with Horizon X, he navigated a narrow pre-accident trail with a steep drop-off. “I never thought I’d trust tech 
like this, but it felt like a knowledgeable friend,” he said. “No overbearing instructions, just calm, accurate 
guidance.”
The device has expanded beyond hiking: low-vision educators lead trips with it, rangers use it for night 
patrols, and rescue teams deploy it in low visibility. Dr. Voss’s team is refining an 80-gram version with 12-hour 
battery life — double the current model — and adding personalized voice commands. “Tech should remove 
barriers, not create them,” she said. “Justin’s journey proves adventure isn’t about sight — it’s about what you 
pursue.”
Cheung, now a Horizon X ambassador, leads monthly hikes for visually impaired people. “Blindness doesn’t 
have to end exploration,” he said. “These glasses aren’t just for trails — they’re for navigating life with courage 
again.”
1．What inspired Dr. Voss to develop the Horizon X?
A．Her hiking hobby
B．A friend’s accident
C．Her grandmother’s illness
D．Blind hikers’ requests
2．What makes the Horizon X stand out from traditional visual aids?
A．It uses AI for real-time analysis
B．It is cheap and easy to carry
C．It works only in outdoor areas
D．It requires no technical skills
3．Why does the author mention Justin Cheung’s first hike with the Horizon X?
A．To show the device’s flaws
B．To prove its practical value
C．To introduce its design team
D．To explain how it was invented
4．What can we infer about the Horizon X from the passage?
A．It will replace human guides soon
B．It has been used by millions of people
C．It is being improved for better performance
D．It was first tested in desert environments
押题5
词数
话题
体裁
考点
押题理由
345
人工智能方法助力狮子保护
说明文
细节理解
推理判断
猜测词义
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
Lions have two types of roar. The newly identified sound is a roar that is mixed in with the recognizable 
“full-throated roar” of the lions. The study is published in the journal Ecology and Evolution.
Researchers from the University of Exeter used artificial intelligence to analyse the roars of animals to 
determine if there was more than one kind of sound. A total of 1,414 roars were analysed during a two-month study 
in the Nyerere National Park in Tanzania, which used microphones hidden in trees. Scientists also obtained more 
than 1,700 roars from collars worn by male lions in the Zimbabwean Bubye Valley Conservancy (BVC). They 
found the roar hidden between grunts (呼噜声) and full-throated roars, which was distinct enough to be classified 
as its own sound. Machine learning methods were able to accurately identify the two roars 95 per cent of the time, 
data show.
“Lion roars are unique and the roaring bout ( 发作) is an important behaviour,” study author Jonathan 
Growcott, a PhD student at the University of Exeter, noted. “Not only does it act as a signal to other lions within 
their pride (狮群) as to where they may be, but, to non-friendly lions, bouts can be used to advertise territorial 
boundaries. Unfortunately, we are no closer to understanding what the communicative differences of the two roar 
types may be. But it has long been considered that lion roars may convey information relating to pride size, age or 
individual.”
Lion populations are dwindling. There are around 25,000 animals in the wild in Africa, around half of the 
number at the start of the century. Mr Growcott added, “Until now, identifying these roars relied heavily on expert 
judgment, introducing potential human bias. Our new approach using AI promises more accurate and less 
subjective monitoring, which is crucial for conservationists working to protect. We believe there needs to be a shift 
in wildlife monitoring and a large-scale change to using passive acoustic ( 声学) techniques. As bioacoustics 
improve, they’ll be vital for the effective conservation of lions and other threatened species.”
1．What is the key discovery reported in Ecology and Evolution?
A．A distinct type of roar.
B．A specific function of roars.
C．A territorial marking method.
D．A newly discovered “full-throated roar”.
2．How did the researchers distinguish the two types of lion roars?
A．By relying on experts’ field observation.
B．By applying AI-based analysis to multiple recordings.
C．By analysing a number of roars collected in one location.
D．By using AI tools to compare roars from different lion prides.
3．What might lion roars act as according to paragraph 3?
A．A means intended to aid hunting.
B．A signature for identifying a specific lion.
C．A tool for expanding the lion pride territory.
D．A way to establish a lion’s status in the pride.
4．What does the underlined word “dwindling” mean in paragraph 4?
A．On the decline.
B．Hard to estimate.
C．Under control.
D．At a stable level.
押题6
词数
话题
体裁
考点
押题理由
306
斯坦福大学研发的一款AI
工具，该工具旨在解决器官
移植中捐赠器官存活时间预
测的难题，提高移植效率
说明文
细节理解
推理判断
猜测词义
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
An AI tool developed by a team of doctors, scientists, and researchers at Stanford University is set to 
transform the field of organ transplantation. As detailed in The Lancet Digital Health journal, this innovative 
technology addresses a long-standing challenge: ensuring donor organs remain alive for transplantation.
Every year, thousands of patients worldwide are stuck on waiting lists for life-saving organ transplants. For 
organs from these donors to be usable, the time between removing life support and the donor’s death must be 
within 30 or 45 minutes — this narrow window is crucial for preserving organ quality. However, relying on 
doctors’ subjective judgment to estimate this time frame has led to significant inefficiencies: about half of all liver 
transplant preparations end up being canceled because the donor does not die within the required window. This not 
only wastes valuable medical resources but also places heavy financial burden on transplant centers.
The new AI tool aims to solve this problem by providing data-driven, accurate predictions. Trained on 
comprehensive data from over 2,000 donors across multiple U.S. transplant centers, the model analyzes key data to 
forecast how quickly a potential donor will progress to death. Most notably, it maintains high accuracy even when 
some donor information is missing, making it highly suitable for real-world clinical settings.
Dr. Kazunari Sasaki, a clinical professor of transplantation, emphasized the tool’s transformative potential. 
“By identifying when an organ is likely to be useful before any preparations for surgery have started, this model 
could make the transplant process more efficient,” he explained, “It also has the potential to allow more candidates 
who need an organ transplant to receive one.” The research team is now looking to expand the tool’s applications, 
with plans to adapt and test it for heart and lung transplants, which could further expand access to life-saving 
treatments for patients in need.
1．What is the function of the AI tool developed by Stanford University?
A．To increase the number of organ donors.
B．To reduce the cost of organ transplant surgeries.
C．To shorten the waiting time for organ transplant candidates.
D．To enhance the accuracy of predicting donor organ survivability.
2．Why are about half of liver transplant preparations canceled?
A．There is no sufficient data to support the AI’s prediction.
B．The death of the donor occurs later than previously estimated.
C．Doctors make mistakes when examining the donor’s organ.
D．The 45-minute window is too short to preserve organ quality.
3．What makes the AI tool suitable for clinical use?
A．It has been tested in transplant centers worldwide.
B．It can be applied to all types of organ transplants.
C．It barely requires donor data to make exact predictions.
D．It maintains precision even with incomplete donor information.
4．Which of the following can be the best title for the passage?
A．An AI Tool for Organ Transplantation B．The Difficulties in Organ Transplantation
C．The Significance of Donor Organ Quality
D．Applications of AI Tool in the Hospitals
三、对人工智能的态度 
押题1
词数
话题
体裁
考点
押题理由
314
英国小说家担忧AI 会取代
其工作、压低收入
说明文
细节理解
推理判断
主旨大意
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
Just over half (51%) of published novelists in the UK say that artificial intelligence is likely to entirely 
replace their work, a new University of Cambridge study has found.59% of novelists say they know their work has 
been used to train AI Large Language Models (LLMs) without permission or payment. Most (85%) novelists 
expect their future income to be driven down by AI. 
The study, led by Dr. Clementine Collett, surveyed 258 novelists to estimate how AI is viewed and used in 
the world of British fiction. Genre authors are considered to be the most easily replaceable, according to the report, 
with 66% of all those surveyed listing romance authors as “extremely threatened”, followed closely by writers of 
thrillers (61%) and crime (60%). 
Despite these concerns, overall attitude in UK fiction is not anti-AI, with 80% of participants in the survey 
agreeing that AI offers benefits to parts of society. In fact, most novelists use AI in their writing process, mainly for 
“non-creative” tasks such as information search. 
Novelists feel copyright (版权) laws have not been respected since the emergence of generative AI. They call 
for respecting copyright and fair payment for the use of their work. Many warn of a potential loss of originality in 
fiction, as well as the loss of trust between writers and readers if AI use is not public. Some worry that suspicion of 
AI use could damage their reputation. 
Writers fear AI may weaken the deep human connection between writers and readers at a time when reading 
is already at historically low levels, particularly among the next generation: only a third of UK children say they 
enjoy reading in their free time. Many novelists want to see more AI-free creative writing on the school curriculum, 
and government-backed initiatives aimed at finding new voices to counter risks of “homogeneity” ( 同质化) in 
fiction brought about by generative AI.
1．What phenomenon does the author point out in paragraph 1?
A．Novelists’ income has stopped growing.
B．Novelists fear AI may change their work.
C．AI brings unexpected benefits to novelists.
D．Novelists actively resist the application of AI.
2．What is the majority view on AI in UK fiction?
A．It can improve creativity capability.
B．Using AI for writing is totally irresponsible.
C．It has no practical value for society.
D．It serves as an aid in the writing process.
3．What is the novelists’ opinion regarding their work being used to train AI?
A．AI training should be banned completely.
B．AI’s innovation level should be enhanced.
C．It should respect their rights and interests.
D．Paying high fees is their top requirement.
4．What can be a suitable title for the text?
A．Will AI Put Novelists Out of Work?
B．How Do Novelists Use AI Tools?
C．Will AI Reshape Our Everyday Life?
D．How Will AI Impact Human Creativity?
押题2
词数
话题
体裁
考点
押题理由
354
人工智能看似进展大，实则
面临可靠性等诸多问题
说明文
细节理解
推理判断
主旨大意
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
To average people, it must seem as if the field of artificial intelligence (AI) is making great progress. Even a 
high-level manager in a high-tech company went so far as to think that the attempt to create AI that has the 
flexibility and resourcefulness of human intelligence — known as artificial general intelligence, or AGI, has 
reached a point where the game is over. Don’t be fooled. Machines may someday be as smart as people and perhaps 
even smarter, but the game is far from over. There is still a large amount of work to be done in making machines 
that truly can comprehend and reason about the world around them. What we need right now is less posturing and 
more basic research. 
At present, AI is making progress — synthetic (合成的) images look more and more realistic, and speech 
recognition can often work in noisy environments — but we are still likely decades away from general-purpose and 
human-level AI that can understand the true meanings of articles and videos or deal with unexpected obstacles and 
interruptions. The field is stuck on precisely the same challenges that academic scientists have been pointing out for 
years: getting AI to be reliable and getting it to cope with unusual circumstances. 
Unfortunately, the fact that these systems still fail to work reliably and struggle with novel circumstances is 
usually buried in the fine print. One system, for instance, worked well on all the tasks the high-tech company 
reported but rarely as well as other contemporary systems. 
Although deep learning has advanced the ability of machines to recognize patterns in data, it has three major 
faults. The patterns that it learns are superficial (表面的), not conceptual. The results it creates are hard to interpret 
and difficult to use in the context of other processes, such as memory and reasoning. As Harvard University 
computer scientist Leslie Valiant noted, “The central challenge going forward is to unify the formulation (规划) of 
learning and reasoning.” Given these limitations, it’s time for AI researchers to concentrate on building systems that 
can learn and reason, not just chase media attention.
1．How does the author develop paragraph 1?
A．By providing statistics.
B．By making assumptions.
C．By supporting the average people’s views.
D．By contrasting the public idea with reality.
2．Which statement best fits the current AI?
A．It is recognized widely.
B．It is facing reliability issues.
C．It can deal with unusual circumstances.
D．It can understand the true meaning of the text.
3．What is the author’s main purpose in writing the last paragraph?
A．To highlight AI’s strengths.
B．To redirect AI research focus.
C．To show AI’s memory processes.
D．To praise the work of AI researchers.
4．What is a suitable title for the text?
A．Treat AI Reasonably
B．AI Should Be Suspected
C．AI Is Making Great Progress
D．Get AI to Recognize Patterns in Data
押题3
词数
话题
体裁
考点
押题理由
349
面对能按需生成整首歌曲的
人工智能（AI）技术时的经
历与思考
议论文
细节理解
推理判断
猜测词义
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
With artificial intelligence (AI) programs that can now generate entire songs on demand, you’d be forgiven 
for thinking AI might eventually lead to the decline of human-made music. As a composer and music educator, I 
happened to start working with the technology.
Music AI can be used in sound analysis, noise removal and mixing. Suno, Beethoven and Udio are some of 
the companies currently leading in the AI music field. In many cases, the outputs don’t have to be excellent; they 
just need to be good enough to compete with real musicians. The music industry is understandably concerned. In 
April 2024, over 200 artists signed an open letter urging AI developers to stop training models with copyrighted (有
版权的) material.
In 2023, I was invited to create a new work with Sydneybased design company Kopi Su, and develop a new 
music AI tool. Would this be a huge waste of time, or end with my data added to some mysterious AI data pool? Or 
would it open up new creative directions for me? Nevertheless, I finally accepted it.
The new tool, called Koup Music, was based on a text-to-image model. For example, I could upload a short 
vocal melody (声乐旋律) and ask the AI to turn it into an insect sound or a“hip hop” style. Though samples were 
limited to 5-10 seconds, they led to unexpected ideas. Some samples were rather pleasing. Others were boring. 
Some came out with unpleasant tones. The imperfection gave me permission to have fun. I focused on generating 
separate musical elements with my text instructions, rather than fully arranged samples.
The process felt like cooperation-like I was making music with a “kooky colleague”. This took away the 
pressure to make “perfect” music, and instead allowed me to focus on new creative possibilities.
The key to a meaningful relationship with music AI is to work alongside it-not let it do all the work. Under 
the right circumstances, it can provide the tools to produce something truly creative. Making “imperfect” art that 
takes time is the price of being human. And I’m grateful for that.
1．Why was the open letter signed in 2024?
A．To seek financial support from AI companies.
B．To encourage musicians to cooperate with AI.
C．To promote the development of AI music tools.
D．To protest the use of copyrighted works in AI 
training.
2．What is the author’s initial attitude towards the work invitation?
A．Curious and eager.
B．Excited and confident.
C．Doubtful and hesitant.
D．Indifferent and uninterested.
3．What can the new AI tool offer to the author?
A．Musical elements with varied quality.
B．Complete songs with sweet melodies.
C．Professional noise removal solutions.
D．Perfectly arranged musical samples.
4．Which of the following might the author agree with?
A．Musicians rely too much on AI tools.
B．AI should be used as a cooperative tool.
C．AI music is a threat to human musicians.
D．The music industry overestimates AI risks.
押题4
词数
话题
体裁
考点
押题理由
323
人工智能在法律领域的应用
现状、优势与局限性
说明文
细节理解
推理判断
主旨大意
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
AI’s application in legal fields is not new — law firms widely use it for document review, like AI-powered 
predictive coding, privilege log recommendations and legal research. Unlike open-source AI that learns from the 
Internet, such tools only rely on provided datasets and instructions.
A 2018 experiment compared AI and experienced lawyers in document review: AI found contract risks with 
94% accuracy in 26 seconds, while lawyers took 92 minutes for 85% accuracy. This shows AI can boost 
productivity. However, worries about losing the “human touch” remain. For example, computer programs that 
assess repeated offenses are praised for data-driven fairness but criticized for possible discrimination and lack of 
empathy, as they may depend on fixed ideas.
Recent focus on open-source AI highlights its limitations: it can make up case information, like fake names 
and dates, since it can’t access relevant case law, endangering lawyers who skip fact-checking. Two cases illustrate 
this: New York lawyers who used ChatGPT for a brief (辩护状) with fake citations (引用) were fined, ordered to 
apologize, and punished; a Colorado associate, pressured for time, submitted an AI-generated brief with fake 
citations, leading to a fine and firing.
Courts have responded with rules. A Northern District of Texas judge requires lawyers to file a certificate 
confirming no generative AI was used in filings, or that AI-drafted content was human-checked via legal databases 
— failure to obey it may lead to Rule 11 sanctions ( 制裁). Hawaii’s federal court demands a “ Reliance on 
Unchecked Source” declaration for AI-used submissions, confirming no fake material.
Critics argue such rules restrict free speech or burden small firms, but most agree AI needs tight control. Law 
firms may adopt policies: requiring human review of AI work, banning AI for legal tasks, and only allowing it for 
non-legal uses like marketing. Given lawyers’ duty of honesty to courts, even accidental misstatements from AI 
may face strict inspection. Therefore, lawyers must use AI cautiously.
1．What do we know about AI’s use in legal fields?
A．AI can correctly predict repeated offenses.
B．AI can be well employed without datasets.
C．AI is applicable in certain aspects of law.
D．AI’s efficiency is far worse than lawyers’.
2．What do the two cases in paragraph 3 show?
A．Using AI is illegal.
B．Examining facts is important.
C．Skipping fact-checks is usual.
D．Submitting citations is reasonable.
3．Why does the author talk about court rules?
A．To demonstrate the abuse of AI.
B．To prove courts’ opposition to AI.
C．To criticize lawyers’ carelessness.
D．To present control measures in AI use.
4．What can be a suitable title for the text?
A．Can Lawyers Trust AI?
B．Can AI Ensure Justice?
C．Can AI Restore the Truth?
D．Can AI Replace Lawyers?
押题5
词数
话题
体裁
考点
押题理由
337
AI 撰写论文引发学术担
忧，研究者开发 xFakeSci 
算法检测AI 论文，该工具
准确率高且将拓展应用领
域，助力维护学术诚信
说明文
细节理解
推理判断
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
Publication is a high aim continuously pursued in academia. But what happens when the publications are 
penned not by professors but by constantly productive AI models that muscle their way into scientific writing?
Large language models like Chat-GPT can produce text by drawing on human writing, raising concerns 
about their misuse in academic circles. In response, Ahmed Abdeen Hamed, a visiting research fellow at 
Binghamton University, has developed xFakeSci, a groundbreaking algorithm that checks AI-generated papers with 
an accuracy of up to 94%.
Hamed’s primary research is in biomedical informatics, and he’s particularly concerned about the 
authenticity(真实性) of information in medical publications and clinical trials. In a study published in Scientific 
Reports, Hamed and Professor Xindong Wu from Hefei University of Technology in China tested xFakeSci using 
150 fake(假的) articles generated by Chat-GPT on topics such as Alzheimer’s, cancer and depression, which were 
compared to an equal number of genuine articles.
The algorithm uncovered distinctive patterns that set apart AI-generated content from human-authored 
papers. One key difference lies in the use of bigrams(二元词组), pairs of words that frequently appear together, 
such as “clinical trials” or “biomedical literature”. AI-generated articles contained fewer unique bigrams but used 
them more frequently throughout the text. This pattern suggests that while AI aims to produce convincing text, real 
scientists focus on accurately reporting their methods and results.
As AI models become more advanced, the line between genuine and fake scientific literature could become 
more unclear. Tools like xFakeSci could become crucial gatekeepers, helping maintain the integrity of scientific 
publications in an age full of AI-generated content.
Despite xFakeSci’s impressive detection rate, Hamed acknowledges that six out of 100 fake papers still slip 
through the net. He plans to expand the tool’s capabilities beyond medicine to other scientific domains and even the 
humanities.
Meanwhile, one thing is clear: The battle against AI-generated fake science is just beginning, but with tools 
like xFakeSci, the scientific community is better equipped to ensure that the pursuit of knowledge remains firmly in 
human hands.
1．What’s xFakeSci’s function based on the text?
A．To write papers.
B．To translate texts.
C．To improve writing.
D．To detect fake papers.
2．What can we know about AI articles?
A．They lack variety.
B．They focus on methods.
C．They use bigrams rarely.
D．They are more convincing.
3．How does Hamed feel about xFakeSci?
A．Critical.
B．Positive.
C．Doubtful.
D．Unconcerned.
4．How is the text developed?
A．By listing data.
B．By giving examples.
C．By making comparisons.
D．By presenting arguments.
四、人工智能与职业
押题
词数
话题
体裁
考点
押题理由
368
人工智能编程
议论文
细节理解
推理判断
猜测词义
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
For decades, computer science students have been taught a central skill: using computers to solve problems. 
In practice, that has meant programming or writing code to tell a machine how to perform tasks.
Now, generative artificial intelligence tools like ChatGPT can write those programs themselves, producing 
code in much the same way they write essays. A student can ask ChatGPT to write a program that sorts a list of 
numbers and get a working answer in seconds. This new capability is transforming how computer scientists get 
work done. Even non-programmers suddenly have the ability to create their own software tools.
This may feel like a threat to some coders’ profession. It certainly marks a significant shift for computer 
scientists and those who educate them. The future of computer science education is to teach students how to master 
the skill of supervision (监管). Why? Because the speed and efficiency of using AI to write code is balanced by the 
reality that it often gets things wrong. These tools are designed to produce results that look convincing, but they 
may still contain errors. There is still a need for humans to play a role in coding — a supervisory one, in which 
programmers oversee the use of AI tools, determine if AI-generated code does what it is supposed to do and make 
repairs to faulty code.
But today’s computer science education still focuses on coding as the primary activity. And worryingly, some 
students are using AI tools to finish their assignments without learning or understanding how the code actually 
works. Most education doesn’t yet emphasize the skills critical for programming supervision. New developers with 
less than a year’s experience can actually be less efficient with AI tools than without, because they are missing 
critical skills and knowledge needed to evaluate and correct what the AI produces.
The rise of generative AI should sharpen our focus on what truly matters in computer science education: 
helping students develop the habits of mind that let them question, reason and apply judgment in a rapidly 
developing field. As the AI pioneer Herbert Simon once advised: “Learning results from what the student does and 
thinks, and only from what the student does and thinks.”
1．What can generative AI tools do in programming?
A．Teach students how to code.
B．Produce code without errors.
C．Write programs following instructions. D．Take the place of experienced coders.
2．The author thinks computer science education should primarily develop students’ ability of ______.
A．supervising and correcting AI-generated code
B．programming or writing code to perform tasks
C．coding or decoding speedily and efficiently
D．understanding and using AI tools independently
3．How does the author support his argument in paragraph 4?
A．By providing statistics.
B．By presenting facts.
C．By clarifying a concept.
D．By making a prediction.
4．What is the author’s purpose in writing this text?
A．To criticize the traditional teaching approaches.
B．To urge a change in computer science education.
C．To emphasize the importance of efficient learning.
D．To show AI tools cannot replace human teachers.
五、人工智能与语言学习
押题
词数
话题
体裁
考点
押题理由
312
人工智能与英语学习结合的
优势、局限性
说明文
细节理解
推理判断
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
The combination of Artificial Intelligence (AI) with English learning is reshaping educational landscapes, 
promising a future of highly personalized instruction. AI-powered platforms can analyze a student’s writing and 
speaking in real time, providing instant, detailed feedback on grammar, vocabulary, and pronunciation — a level of 
attentiveness impossible in a traditional classroom of forty students. This allows learners to practice and progress at 
their own pace. A 2025 study by the University of Oxford found that students using an AI conversation tutor for 30 
minutes daily showed significantly greater fluency gains over eight weeks compared to a group relying only on 
textbook exercises.
However, this technological promise is not without its pitfalls. Critics argue that an over-reliance on AI may 
lead to a narrow, stereotypical view of language. Language is more than a set of rules to be mastered; it is a 
fundamentally human tool for cultural exchange, emotional expression, and critical thinking. An AI can correct a 
misused idiom, but it cannot yet grasp the subtle idea in a novel behind the lines or engage in an instinctive debate 
about its themes. The risk is that learners, while becoming technically accurate, might fail to develop the deeper 
cultural understanding and creative flexibility that come from human interaction.
The most productive path forward, therefore, lies not in choosing between human teachers and AI, but in 
strategically combining them. Imagine a classroom where an AI handles routine drills and personalized review, 
freeing the teachers to let them conduct group discussions, project-based learning, and activities that build 
creativity and productivity ability. In this model, the teacher shifts from a sole knowledge-deliverer to an advisor or 
guider, helping students to apply their AI-practiced skills in rich, thoughtful contexts. The ultimate goal is a 
partnership where technology handles precision and personalization, while humans focus on inspiration, 
connection, and the higher-order thinking that makes language truly come alive.
1．According to the Oxford study, what was a key finding about using an AI conversation tutor?
A．It reduced students’ anxiety about making mistakes.
B．Students using it showed faster improvement in spoken fluency.
C．It was most effective when used for more than one hour per day.
D．Its feedback was considered more motivating.
2．What is a major limitation of current AI in English learning according to paragraph 2?
A．Its inability to provide feedback.
B．Its high cost which makes it inaccessible for most students.
C．Its lack of capacity for understanding deeper cultural and emotional language.
D．Its tendency to provide wrong corrections for complex sentences.
3．What can be inferred about the future role of human teachers?
A．They will become outdated as AI technology continues to advance.
B．Their role will change towards tasks that require human creative and productive thinking.
C．They will need to focus more on delivering knowledge.
D．Their importance will decrease to merely guiding students’ use of AI tools.
4．What is the main purpose of the passage?
A．To advertise a learning application.
B．To argue against the use of any technology in traditional language classrooms.
C．To present and advocate a balanced perspective to combine AI with English learning.
D．To show the advantages of language learning by AI.
六、人工智能的缺陷
押题1
词数
话题
体裁
考点
押题理由
383
社交AI 无法提供真实的人
类认可与理解
说明文
细节理解
推理判断
主旨大意
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
The rapid evolution of artificial intelligence (AI) is sparking intense debate about its role in human 
relationships and emotional well-being. While Bill Gates predicted AI could replace educators and healthcare 
workers, others envision a future where humans befriend and even fall in love with AI. This shift raises profound 
questions about the nature of social connection and what it means to truly relate to another being.
Sociologist Allison Pugh highlights a subtle ( 微妙的) yet critical concern: the loss of authentic human 
recognition. She recalls observing a boy in a remote online school who, after solving a math problem, looked 
around for someone to share his success with. No one was watching. Advanced social AI can pretend to “pay 
attention to you”, but Pugh argues this is a different experience. Before cell phones, a phone call might connect you 
with an entire household, building up a sense of community. Now, technology forces us to view each other as 
isolated individuals, breaking our social lives into small parts and making our connections more like a deal.
Pugh warns that social AI, being programmable and lacking an independent perspective, offers a “false” kind 
of friendship. Unlike human relationships, which involve mystery, challenge, and the acceptance of an “other,” AI 
is an extension of ourselves. It can be tailored to our preferences but cannot truly understand or challenge us. Real 
connection requires genuine understanding — an opposing perspective that AI lacks. As human-like AI competes 
for our attention, we must consider what is lost when we replace the complexity of human interaction with the 
convenience of a machine. The convenience of AI may come at the cost of our deepest human needs for mutual (相
互的) understanding.
This distinction is crucial because human growth often occurs through friction ( 摩擦). When we interact 
with someone who has a different background, opinion, or emotion, we learn empathy (共情) and adaptability. An 
AI companion, designed to please and agree with us, removes this friction, potentially holding back our emotional 
development. We risk becoming isolated in a bubble of our own preferences, losing the ability to navigate the 
messy, unpredictable, but ultimately rewarding dynamics of real human relationships. Therefore, while AI can 
provide comfort, it cannot replace the unique value of a bond forged between two autonomous beings.
1．Why does Pugh mention the boy studying online?
A．To highlight the rapid evolution of AI.
B．To contrast past and present social ties.
C．To stress the forces of advanced technology.
D．To show the problem of lacking human acknowledgement.
2．What can we infer about AI friendship?
A．It has independent thinking ability.
B．It fails to bring true understanding.
C．It contains mystery and challenge.
D．It can promote human adaptability.
3．What causes human emotional development?
A．Convenience from advanced machines. B．Common preferences among people.
C．Disagreement from varied human interactions.
D．The comfort provided by AI companions.
4．Which is the best title for the passage?
A．AI: A Threat to Human Relationships
B．AI: An ideal substitute for Human Recognition
C．AI: A New Way to Connect with OthersD．AI: A New Trend for Friendship and Love
押题2
词数
话题
体裁
考点
押题理由
331
人工智能存在不可靠和不准
确问题
说明文
细节理解
推理判断
猜测词义
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
One of the world’s biggest tech companies could see AI slowing its roll after the new report exposed just 
how careless and alarmingly inaccurate the technology in its current form can really be. If you’re thinking AI is a 
fast track to handing in your homework, you might want to think again. 
A striking report from Apple questions AI’s reliability and potential. Researchers tested large reasoning 
models used in platforms like DeepSeek and Claude and large language models (the basis of ChatGPT). Both 
performed poorly in complex tasks, though language models did better in simple ones. Notably, reasoning models 
“reduced their effort” when tasks got too challenging, which the report called “particularly concerning”. 
Problems with these models aren’t new. OpenAI claimed its o3 and o4-mini models were its “smartest,” but 
researchers found o3 was wrong 51% of the time, and o4-mini 79%. Apple stopped its AI-powered news alerts after 
major mistakes, such as a false claim about tennis star Rafael Nadal and an early announcement of a darts 
championship winner. The BBC found 51% of AI answers on news had “significant issues”, with 19% of those 
involving its content being factually wrong, and 13% of quotes from its stories altered or made up. A Chicago 
newspaper once published an AI-generated summer reading guide with non-existent books, and a Dublin 
Halloween parade advertised online was fake. Google’s AI also gave odd suggestions like adding glue to pizza. 
Experts like Dr. Niousha Shafibady note expecting AI to be a “magic wand” is wrong. AI fails in complex 
tasks due to algorithms’ (算法的) essential characteristics — they may lose logical reasoning, miss data patterns, or 
waste resources on wrong solutions. 
While companies say AI is improving, experts disagree. A survey by the Association for the Advancement of 
Artificial Intelligence found 60% of respondents don’t think AI’s accuracy and trustworthiness issues will be solved 
soon. Its president, Francesca Rossi, stressed the need to develop AI responsibly, ensuring it supports humanity and 
matches human values.
1．What is one problem of reasoning models according to Apple’s report?
A．They can refuse simple assignments.
B．They can put less energy into hard work.
C．They can weaken language model functions.
D．They can cause mistakes in platform operations.
2．Why are the Chicago newspaper and Dublin parade mentioned?
A．To show misinformation of AI.
B．To give suggestions on AI errors.
C．To prove the output ability of AI.
D．To blame media for fake AI contents.
3．What does “magic wand” in Paragraph 4 probably refer to?
A．A special data tool.
B．A real magic stick.
C．An essential language model.
D．A perfect problem-solver.
4．What can we learn from the passage?
A．Experts are satisfied with AI’s improvement.
B．Companies will slow developing AI for its reliability.
C．People should be cautious using AI for accurate tasks.
D．Most respondents are confident about AI’s humanity.
押题3
词数
话题
体裁
考点
押题理由
321
过度依赖外部工具（尤其是
人工智能）对人类认知能力
的潜在负面影响
说明文
主旨大意
推理判断
猜测词义
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
Cognitive (认知的) offloading, the act of delegating mental work to external tools, should concern anyone 
who values their own intelligence. When given the choice between thinking through a problem or letting 
technology handle it, will we choose the tool, even when we’re perfectly capable of succeeding on our own?
Sam Gilbert noticed in his experiments something he calls “reminder bias.” This is the tendency to use 
external memory aids even when your own memory would serve you better. Participants chose to set digital 
reminders for tasks they could easily remember unaided. Even when offered money to encourage people to rely on 
their own memory, they still couldn’t overcome it.
The root reason seems to be that people underestimate their own cognitive abilities. They think they’ll forget, 
so they set a reminder. They think they can’t solve the problem, so they ask AI. Over time, by consistently choosing 
the external tool, they never exercise the internal capability.
If the problem were limited to memory, it might not matter much. After all, what’s the problem with setting a 
reminder on your phone? But research suggests the cognitive costs extend far beyond remembering appointments.
Nathaniel Barr claimed that individuals who score lower on the Cognitive Reflection Test show greater 
smartphone dependence when obtaining information. These are people who, when faced with a problem, 
instinctively (本能地) reach for the quickest solution.
AI hasn’t created this tendency. Humans have often preferred mental shortcuts. But AI has removed nearly 
all difficulties. You don’t have to go to a library or even type a search question. You just ask, and the answer 
appears, exactly as you need it.
My concern is that people may begin using AI in ways that prevent the development of analytical skills. For 
a college student who consistently uses AI to work through problems, I question whether they’re building the neural 
infrastructure (神经基础结构) to handle similar problems independently later in life.
1．What does the underlined word “delegating” most probably mean?
A．Assigning.
B．Comparing.
C．Applying.
D．Linking.
2．Why did participants in Gilbert’s experiment set unnecessary digital reminders?
A．They lacked confidence in their own memory.
B．They were not offered enough money.
C．They were poor in cognitive abilities.
D．They never exercised their internal capability.
3．The author mentions the Cognitive Reflection Test to show ______.
A．people rely too much on smartphones
B．AI works better than human memory
C．reminder bias causes harm to more than just memory
D．smartphones are the quickest solution to many problems
4．What is the best title of the article?
A．Cognitive Offloading: The Cause of AI Reliance
B．How External Aids Like AI Enhance Reminder Bias
C．External Tools or Internal Tool,Which Causes More Harm
D．Over-reliance on External AI Tools May Weaken Cognitive Abilities
押题4
词数
话题
体裁
考点
押题理由
333
人工智能系统可能永远无法
真正实现安全
说明文
细节理解
推理判断
主旨大意
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
Artificial intelligence inspires a mix of wonder and anxiety. It promises efficiency, creativity, and economic 
growth. But as its use spreads, an uncomfortable truth emerges: AI systems may never be truly secure. Their 
security weaknesses are not minor errors but the unavoidable consequence of how these technologies are built. 
Traditional software has boundaries: defined inputs, formal reasoning, predictable outputs. Security relies on 
spotting errors and fixing them. AI is different. Models learn from oceans of data, forming patterns even their 
designers cannot fully trace. This cloudiness — praised as intelligence — also makes assurance impossible. What 
cannot be understood cannot be secured. 
Opponents thrive in that uncertainty. Slightly changed images fool facial recognition. A handful of poisonous 
data points influence training results. Carefully crafted prompts (提示词) tempt systems to reveal secrets. Unlike 
conventional attacks, these weaknesses cannot be patched away ( 补丁修复). They are woven into the learning 
process itself, and therefore permanent. 
The risks rise with adoption. A chatbot talking nonsense is an annoyance. An AI in charge of medical 
diagnoses or financial trades is another matter. Imagine patient outcomes damaged by poisonous inputs, or stock 
markets shocked by engineered algorithms. When fragile systems run critical infrastructure, fragility transforms 
into systemic risk. 
Defenders are not unoccupied. Opposing training, model monitoring, and data sanitation ( 净化) aim to 
bounce back. Yet each solution has limits. Training against one attack leaves blind spots for another. Security layers 
add costs and slow performance — burdens unwelcome in a global AI arms race that rewards speed over caution. 
Like nuclear power or aviation, AI may require ongoing monitoring, legal framework, and international 
norms. Yet regulatory responses are often reactive, lagging far behind technological leaps. 
History suggests severe lessons. The Internet was built without security at its core, and decades later society 
still struggles with hacking, disinformation, and cybercrime. AI seems certain for the same path: constant security 
weaknesses, partial fixes, and a permanent arms race between attackers and defenders. Perfection will remain out of 
reach.
1．Why is AI security fundamentally different from traditional software?
A．AI systems use formal reasoning for reliable outputs.
B．Traditional software has much more minor errors to fix.
C．AI’s security weaknesses are easier to patch through updates.
D．Traditional software has predictable and traceable boundaries.
2．What does the author think of the current efforts to defend AI systems?
A．They are sufficient to address all potential threats.
B．They are the main driver of the global AI arms race.
C．They cannot be fully secured without a balancing act.
D．They have kept up with the speed of AI development.
3．Why does the author mention the history of the Internet?
A．To predict a future of perfectly secure AI.
B．To present a security model for AI to follow.
C．To claim regulation can easily prevent AI risks.
D．To imply AI security may be a lasting social issue.
4．Which of the following is the best title for the text?
A．The Bright Future of AI: Efficiency and Creativity
B．The Security Dilemma of AI: A Built-in Challenge
C．The Path to Perfect Security: Building Unbreakable AI
D．The Weaknesses of AI’s Security: A Temporarily Challenge
押题5
词数
话题
体裁
考点
押题理由
356
AI 垃圾内容的特征与制作
目的，指出其泛滥挤占优质
内容空间、损害创作者利益
说明文
细节理解
推理判断
猜测词义
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
You’ve probably encountered images in your social media feeds that look like a cross between photographs 
and computer-generated graphics. Some are imaginative and some are believable at a quick look — remember the 
little girl carrying a little dog in a boat during a flood? These are examples of AI slop ( 垃圾), low-to mid-quality 
content — video, images, audio, text or a mix — created with AI tools, often with little regard for accuracy. It’s 
fast, easy and inexpensive to make this content. AI slop producers typically place it on social media to employ the 
economics of attention on the Internet, displacing higher-quality material that could be more helpful.
AI slop has been increasing over the past few years. As the term “slop” indicates, that’s generally not good 
for people using the Internet. The Guardian published an analysis in July 2025 examining how AI slop is taking 
over YouTube’s fastest-growing channels. The journalists found that nine out of the top 100 fastest-growing 
channels feature AI-generated content. In many cases, people post AI slop that’s just good enough to attract and 
keep users’ attention, allowing them to profit from platforms that monetize ( 变现) streaming and view-based 
content.
The ease of generating content with AI enables people to post low-quality articles to publications. 
Clarkesworld, an online science fiction magazine that accepts user posts and pays contributors, stopped taking new 
posts in 2024 because of the flood of AI-generated writing it was getting.
AI slop also harms artists by causing job and financial losses and crowding out content made by real 
creators. The placement of this lower-quality AI-generated content is often not distinguished by the algorithms (算
法) that drive social media consumption, and it displaces entire classes of creators who previously made their 
livelihood from online content.
Wherever it’s enabled, you can flag content that’s harmful or problematic. On some platforms, you can add 
community notes to the content to provide context. For harmful content, you can try to report it. Along with forcing 
us to be on guard for “inauthentic” social media accounts, AI is now leading to piles of rubbish degrading our 
media environment.
1．Why do producers create AI slop according to Paragraph 1?
A．To replace professional materials.
B．To improve artistic expression globally.
C．To use Internet attention economies for profit.
D．To educate Internet users about new technologies.
2．What can we infer from Clarkesworld’s action in 2024?
A．Science fiction became less popular.
B．Poor AI content flooded their system.
C．Their subscriber base doubled unexpectedly.
D．They began paying higher fees for human writers.
3．What should individuals do regarding the rise of AI slop?
A．Stop using AI-generated content.
B．Avoid engaging with social media platforms.
C．Rely on algorithms to screen out low-quality content.
D．Take action by marking harmful AI-generated content.
4．Which of the following would be the best title for the text?
A．AI Slop: A New Form of Art
B．The Positive Effects of AI on Media
C．The Role of Social Media in Promoting AI Slop
D．AI Slop: The Growing Threat to Online Content
七、规范人工智能的发展
押题
词数
话题
体裁
考点
押题理由
346
制定通用的AI 治理框架与
规范
说明文
细节理解
推理判断
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
China on Wednesday submitted a position paper on strengthening the ethical governance ( 伦理治理) of 
artificial intelligence (AI) to the 2022 Meeting of the High Contracting Parties to the Convention on Certain 
Conventional Weapons held in Geneva. 
Li Song, China’s ambassador for disarmament ( 裁军) affairs, told the participants that China finds it 
important to enhance the understanding of all countries on AI ethics and to ensure that AI is safe, reliable, 
controllable and capable of better empowering global sustainable development and enhancing the common well-
being of all mankind. 
“As the most representative disruptive technology, AI, while providing enormous potential development 
benefits to human society, has also brought uncertainty that may give rise to multiple global challenges and even 
fundamental ethical concerns,” Li said. 
The position paper China submitted on Wednesday suggests that governments should give priority to ethics, 
establish and improve rules, norms and accountability mechanisms for AI ethics, clarify responsibilities and power 
boundaries of AI-related entities, fully respect and protect the legitimate rights and interests of all groups, and 
respond to relevant ethical concerns at home and abroad in a timely manner. 
According to the paper, governments should prohibit the use of AI technologies and relevant applications 
that run counter to laws, regulations, ethics and standards. They should also strengthen quality monitoring and 
evaluations on the use of AI products and services, and formulate emergency mechanisms and compensation 
measures. 
The position paper also urges governments to encourage transnational, interdisciplinary ( 跨学科的), and 
cross-cultural exchanges and cooperation, ensure that the benefits of AI technologies are shared by all countries, 
promote the joint participation of countries in international discussions and rules-making on major issues regarding 
AI ethics, and oppose the building of exclusive groups and malicious obstruction ( 恶意阻挠) of other countries’ 
technological development. 
“China calls on the international community to reach an international agreement on the issue of AI ethics on 
the basis of broad participation, and to work to formulate widely accepted international AI governance frameworks, 
standards and norms while fully respecting the principles and practices of different countries’ AI governance,” the 
paper concludes.
1．What is the primary purpose of China’s position paper on AI?
A．Compete for AI weapons.
B．Ensure AI safety and ethics understanding.
C．Seek global AI leadership.
D．Pursue international AI cooperation.
2．What does paragraph 3 imply?
A．AI can be double-edged.
B．AI boosts social progress.
C．The ethical problem is very serious.
D．Human society is developing quickly.
3．What should governments do for AI ethics?
A．Leave it to international organizations. B．Focus only on fast AI development.
C．Give priority and set rules.
D．Allow free AI use in all fields.
4．What is China’s attitude to international AI ethics governance?
A．Isolated and indifferent.
B．Passive and conservative.
C．Active and cooperative.
D．Critical and skeptical.
八、人工智能与医疗发展
押题1
词数
话题
体裁
考点
押题理由
324
中医AI 辅助研究的相关情
况
说明文
细节理解
推理判断
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
When two different traditional Chinese medicine (TCM) physicians see the same patient, they can have 
opposing views on the patient’s diagnosis and treatment plan. To address this subjectivity problem, Kon Wen Xuan, 
a recent graduate from Nanyang Technological University (NTU) in Singapore with a double degree in Biomedical 
Sciences & Chinese Medicine, started an AI research project. The project aimed to guide practitioners, particularly 
students and new professionals, on how to deal with diseases they are unfamiliar with.
“So a physician might just input the symptoms and the AI can immediately predict a diagnosis and suggest a 
prescription,” Wen Xuan explains.
Wen Xuan processed patient data from the NTU TCM clinic and used it to train machine-learning models. 
He processed 50,000 records, of which about 9,000 were usable. Transforming the raw information into a system 
readable by AI was a time-consuming task. He carefully categorized symptoms to maintain consistency in data 
interpretation.
The AI was better at answering simple questions like whether the patient is “heaty” and predicting simple 
disease sub-categories. But it faced difficulties when things got more complex. Its reliance on the limited data it 
was fed also posed significant challenges. The diversity of conditions in the data, totaling around 130 diseases, also 
raised concerns about whether there was enough data for certain illnesses.
Considerations of AI use are crucial. “Who will be responsible if a patient is wrongly diagnosed?” Wen Xuan 
asks. He envisions the AI tool as a supportive reference to aid students and new TCM professionals, rather than as 
replacement for practitioners. He also emphasizes the need for standardization in training AI. “An expert group 
should decide what data to use and how to train the AI,” says Wen Xuan. 
Wen Xuan plans to enhance his AI research, believing there is interdisciplinary potential for the research. 
“Western medicine practitioners can potentially use the AI tool to understand how TCM works and also do research 
in this area,” he explains.
1．Why did Wen Xuan conduct the project?
A．To develop a new type of medicine.
B．To create a database for patient records.
C．To provide training for TCM students.
D．To help TCM practitioners with informed advice.
2．What was the problem when using AI for the project?
A．It failed to provide quick responses.
B．It struggled with complex conditions.
C．It had difficulty in simplifying the answer.
D．It was unable to process raw information effectively.
3．What does Wen Xuan suggest about standardizing AI training?
A．Getting an expert team to offer guidelines.
B．Enabling the AI to collect information online.
C．Focusing the training on the most common diseases.
D．Training the AI with data from experienced practitioners.
4．What is a possible benefit of Wen Xuan’s AI research?
A．AI could replace some Western medicine practices.
B．AI could be used to perform challenging operations.
C．Western medicine industry could be advanced by AI.
D．Western doctors could employ AI to understand TCM.
押题2
词数
话题
体裁
考点
押题理由
303
虚拟现实技术从娱乐用途转
向医疗领域并带来变革的相
关情况
说明文
细节理解
推理判断
猜测词义
主旨大意
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
What sounds like science fiction is becoming an increasingly common reality in medical facilities 
worldwide. Originally, virtual reality (VR) glasses, primarily associated with gaming and entertainment, are now 
emerging as powerful medical tools. Whether applied in physical recovery, pain management, or surgical training, 
this wearable technology is transforming traditional medical practices. Patients can be lost in interactive therapeutic 
environments while receiving real-time guidance for their treatment.
The advantages of medical VR systems begin with their user-friendly design. Modern VR wearable devices 
feature lightweight construction and wireless connectivity, ensuring comfortable extended use. High-definition 
displays create realistic virtual environments, while integrated sensors precisely track body movements. Advanced 
models even combine eye-tracking technology to monitor patients’ visual attention, providing valuable data for 
healthcare professionals. Beyond creating realistic experiences, these systems employ complicated software to 
enhance treatment outcomes. Artificial intelligence algorithms (算法) can analyze patients’ movements in real time, 
automatically adjusting therapy difficulty to match individual progress. The combination with cloud technology 
allows for secure data storage and enables remote monitoring by medical teams. This connectivity also lets different 
systems work together by connecting with other digital health platforms, like telemedicine services and wearable 
fitness trackers.
The practical benefits of VR in healthcare are epoch-making. For patients, interactive virtual environments 
increase engagement and motivation during recovery exercises. For medical institutions, these systems can reduce 
staff workload while providing detailed progress tracking. Some VR systems have already obtained official medical 
device certification, confirming their safety and effectiveness for clinical use.
Looking ahead, as technology continues to advance, VR applications in medicine are expected to expand 
further. Current research explores their potential in treating psychological conditions, managing chronic pain, and 
enhancing medical education. While not replacing traditional methods entirely, VR technology offers a valuable 
supplement to traditional treatments, potentially making healthcare more accessible and effective for patients 
everywhere.
1．What was VR mainly used for in the beginning?
A．Leisure activities.
B．Medical practices.
C．Remote monitoring.
D．Interaction guidance.
2．What can be inferred about medical VR systems from paragraph 2?
A．They merely rely on visual effects to work.
B．They aim to adapt treatment to each patient.
C．They disconnect with other medical platforms.
D．They require constant manual control by doctors.
3．What does the underlined word “epoch-making” in paragraph 3 probably mean?
A．Official.
B．Theoretical.
C．Substantial.
D．Controversial.
4．What does the last paragraph mainly say about modern VR wearable devices?
A．Their origin.
B．Their definition.
C．Their prospects.
D．Their shortcomings.
押题3
词数
话题
体裁
考点
押题理由
350
中医舌诊智慧与人工智能技
术结合
说明文
细节理解
推理判断
主旨大意
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
Combining ancient Chinese medical wisdom with cutting-edge artificial intelligence (AI), researchers have 
developed a computer algorithm (算法) that predicts a wide range of diseases with an impressive 98% accuracy by 
analyzing the color, shape, and coating of the human tongue. 
For thousands of years, traditional Chinese medicine (TCM) has regarded tongue examination as a core 
diagnostic tool, holding that the tongue serves as a “window to internal health” that reflects the body’s physical 
changes. Today, researchers from Middle Technical University (MTU) and the University of South Australia 
(UniSA) have modernized this traditional practice through advanced machine learning technology. 
The AI model went rigorous training on 5,260 high-resolution tongue images and was tested on 60 clinical 
samples from patients with confirmed health conditions, guaranteeing its reliability across diverse cases. The 
system employs specialized cameras positioned 20 centimeters from the patient to capture subtle tongue features in 
real-time, without the need for invasive procedures. As senior author Adjunct Associate Professor Ali Al-Naji notes, 
distinct tongue characteristics-color, shape, and coating thickness-can reveal specific health issues: diabetes patients 
often present with a yellow tongue, cancer patients with a purple tongue and thick greasy coating, and acute stroke 
patients with an unusually shaped red tongue. 
Co-author UniSA Professor Javaan Chahl emphasizes the technology’s unique advantages: it is secure, 
efficient, user-friendly, and affordable, acting as a valuable complement to modern disease screening methods. The 
system has demonstrated great promise in diagnosing over common conditions, including diabetes, stroke, anemia, 
asthma, liver and gallbladder disorders, and COVID-19. 
By integrating TCM’s overall insights with AI’s powerful data-processing capabilities, it enables quick, 
accurate, and accessible health screenings, which may in turn lead to earlier disease detection, improved patient 
outcomes, and reduced healthcare costs. Furthermore, this innovation explores the meaningful intersection of 
traditional and modern medicine, showing how ancient practices can enhance contemporary healthcare and promote 
more culturally inclusive medical systems. 
The development of this technology also raises interesting questions about the intersection of traditional and 
modern medicine. It demonstrates how ancient practices can inform and enhance contemporary medical 
approaches, potentially leading to more overall and culturally inclusive healthcare systems.
1．What can we learn about the integration of AI and TCM?
A．It creates a brand-new diagnostic theory.
B．It replaces TCM with modern technology.
C．It upgrades TCM tongue diagnosis with AI.
D．It overlooks the role of tongue examination.
2．How does the AI system ensure the reliability of its diagnostic results?
A．By testing on clinical samples
B．By adopting invasive procedures
C．By relying on subjective judgment
D．By using general cameras to capture images
3．Which can be inferred about the integration of TCM and AI?
A．It is a primary screening tool.
B．It improves screening accessibility.
C．Modern medicine ignores cultural diversity.
D．It cures diseases through early detection.
4．What is the main idea of the passage?
A．TCM is superior to modern medicine.
B．AI cures various common diseases.
C．Tongue color is key to disease detection.
D．AI transforms TCM tongue diagnosis.
押题4
词数
话题
体裁
考点
押题理由
326
AI 如何“再人性化”医疗
说明文
细节理解
推理判断
猜测词义
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
In a conversation with Drs. Ben Ebert and Eliezer Van Allen of Dana-Farber Cancer Institute, Van Allen 
said,“AI will re-humanize medicine.” This phrase captured the central issue: modern health care in the United 
States often feels inhuman, dominated by paperwork, disputes, and rigid procedures, despite the best intentions of 
medical professionals.
Over the past 25 years, digitization of medical records has not improved patient experience or clinical 
outcomes. Dr. Allen Chang noted that electronic health records (EHRs) have failed to enhance job satisfaction or 
operational efficiency. The 2021“quadruple aim” paper by Microsoft and University College London outlined goals 
to improve population health, patient experience, caregiver satisfaction, and reduce costs. Digitization alone has 
fallen short, but AI offers potential to make care more personalized, precise, predictive, and accessible.
AI is already transforming doctor-patient interactions. “Ambient scribes” transform spoken consultation into 
notes for EHRs, allowing physicians to engage directly with patients instead of typing, improving satisfaction and 
reclaiming time previously spent on documentation. AI also aids diagnostic ( 诊断的) thinking, supplementing 
critical reasoning that doctors rarely have time for, acting as a “House on demand” with vast medical knowledge.
In research, AI accelerates drug discovery. Alpha Fold predicted structures for millions of proteins, vastly 
more than previously possible. AI also helps analyze literature, repurpose drugs, and match patients to clinical 
trials, increasing efficiency and enabling better decisions. Human expertise remains essential, as AI models must 
understand disease complexity, lab-specific biases, and biological interactions.
Looking forward, AI can automate workflows, predict disease evolution, monitor patients remotely, and 
support virtual care. Combining clinician insight with AI promises to re-humanize health care while democratizing 
access. Ebert and Van Allen anticipate fully AI-enabled hospitals where patient monitoring, personalized 
therapeutics, and research integration create a smooth, humane experience.
In short, AI does not replace doctors — it restores their ability to connect with patients supports complex 
decision-making, and enhances care quality, fulfilling the promise of health care that is truly personal and human-
centered.
1．Why does the author mention EHRs in the second paragraph?
A．To explain a solution.
B．To make a comparison.
C．To introduce a problem.
D．To present an argument.
2．What does the underlined phrase “Ambient scribes” in paragraph 3 refer to?
A．Databases that store medical records.
B．Software that records clinical interactions.
C．Devices that improve patients’ well-being.
D．Notes that summarize physicians’ diagnoses.
3．What do Ebert and Van Allen focus on regarding AI-enabled hospitals?
A．A clinical study.
B．A statistical analysis.
C．A practical application.
D．A theoretical framework.
4．What is the author’s attitude toward AI-enabled hospitals?
A．Unclear.
B．Disapproving.
C．Doubtful.
D．Favorable.
押题5
词数
话题
体裁
考点
押题理由
359
AI 完全取代医生的观点，
指出AI 局限性，主张人机
协作推动医疗发展
议论文
细节理解
推理判断
猜测词义
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
As a general doctor and medical educator deeply committed to doctor-patient relationships, I approach 
Charlotte Blease’s Dr Bot with an open mind toward AI’s potential in healthcare. While I fully acknowledge AI as a 
valuable innovative tool that can improve medical efficiency, the book’s core argument that AI should completely 
replace doctors made me speechless. Blease’s perspective overlooks the great advancement of modern medical 
practice and the irreplaceable human elements that lie at the heart of effective care.
Her reasoning contains critical mistakes: she ignores the importance of caring for others — a quality 
extensively proven to boost patient satisfaction, treatment following and overall clinical outcomes — as an obstacle 
to sensible clinical decision-making. She also ignores AI’s natural limitations: it cannot perform physical 
examinations, surgeries or handle complex emergency situations that demand human judgment and adaptability. 
Additionally, she underestimates pressing risks associated with AI, including data leaks, unequal access to 
advanced tools across different populations and the too much energy and resource consumption tied to AI 
operations.
AI already contributes meaningfully to medicine, from making easier notes about communicating with 
doctors to speeding up literature reviews, and Australian doctors readily accept these innovations to reduce 
administrative workload. Younger doctors, raised in the digital era, are particularly eager to integrate AI into their 
daily practice. Yet as physician Dhruv Khullar warns, over-reliance on AI risks diminishing doctors’ diagnostic 
(诊断) skills and clinical judgment due to a lack of practice, leaving patients in danger when technology fails — an 
increasingly likely situation during climate-driven problems and system failures.
I openly acknowledge the imperfections of the medical profession, including occasional instances of 
overconfidence, insensitivity and human error. However, today’s doctors — especially younger generations — 
bring greater awareness of diverse patient needs, strong communication abilities and a genuine commitment to 
using AI to improve care quality. The path forward doesn’t lie in replacing doctors with AI, but in building good 
teamwork between the two. Blease’s call for full doctor replacement is shortsighted. True healthcare progress 
depends on using AI to boost, not replace, the human kindness, clinical skills, knowledge, and contextual 
understanding — the qualities that define effective, patient-centered medicine.
1．Which is a noted shortcoming of AI in medicine?
A．Lack of data processing ability.
B．Inability to conduct surgeries.
C．Refusal to integrate with clinics.
D．Failure to analyze medical papers.
2．What does the underlined word “diminishing” in paragraph 3 probably mean?
A．Weakening.
B．Deepening.
C．Refreshing.
D．Demonstrating.
3．What is the author’s argument?
A．Banning AI in clinical settings.
B．Fully replacing doctors with AI.
C．Promoting AI-doctor cooperation.
D．Limiting AI to administrative work.
4．What is the author’s attitude toward Blease’s core argument?
A．Fully supportive.
B．Strongly critical.
C．Mildly doubtful.
D．Totally indifferent.
九、人工智能与农业发展
押题
词数
话题
体裁
考点
押题理由
421
探讨了人工智能在农业领域
的应用及其影响
说明文
细节理解
推理判断
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
On farms all over the country, even the most traditional ones, thought is being given to the possibilities of AI. 
For the moment, artificial intelligence isn’t making much of an impact on the daily lives of most British farmers. 
But the way AI can change agriculture in the near future is very much a reality in universities and research labs. My 
eyes were opened to the possibilities on a visit to a lab where cows were being automatically (自动化) monitored in 
an innovative ‘smart’ dairy. AI was learning to recognize each individual cow and keep a record of its behavior. It 
then analyzed the data to warn of any signs of illness. 
Meanwhile, in my home county, work is underway to ensure that the upcoming generation of farmers isn’t 
left behind. In a few weeks’ time, newly arrived students at the Royal Agricultural University (RAU) will start their 
course in Agricultural Technology, at the heart of which is AI’s role in making farming more effective, productive 
and adaptive in a world of changing weather patterns. This would have sounded like science fiction in the 1950s, 
but now, 70 years on, I’m encouraged that tomorrow’s farmers are being armed with the know-how to succeed in 
the new digital age. 
Naturally, there are concerns about what all this might mean for the countless number of people who depend 
on farming for their livelihood (生计). Soft fruit and seasonal vegetables are extremely tender and need an army of 
pickers and packers who may find themselves replaced at some point by the sort of standard precision-robots that 
have been trialled (试用) on a strawberry farm. 
In recent years many places have struggled to employ and keep seasonal workers, so rather than create 
unemployment, AI might just provide the answer to a worrying labor gap. It’s worth saying that this technology will 
need a whole new set of workers to produce the hardware, input new data and then install (安装), maintain, repair 
and replace systems and kit. So don’t lose hope but do consider upskilling (技能提升). 
Since the Industrial Revolution, every generation of farmers has had to deal with the technological changes 
of the day and they have successfully adapted to survive. Who knows! AI may well turn out to be just the latest in a 
long line of challenges to be accepted by adaptable and spirited farmers. And anything that saves me the worry of 
checking the weather forecast day and night during harvest season must be a good thing.
1．What did the author discover during his visit to the lab?
A．AI has made automation a reality in the dairy industry.
B．AI has helped most British farmers to closely monitor cows.
C．AI is being developed to improve cow health through behavior analysis.
D．AI is primarily being used to change universities and labs across Britain.
2．What is the focus of the course in Agricultural Technology at the RAU?
A．The role of traditional methods in increasing crop productivity.
B．The development of AI in agricultural practices in the UK.
C．The ways climate change affects global food security.
D．The impact of AI on improving farming efficiency and productivity.
3．What can be inferred about the introduction of AI in agriculture from the third and the fourth paragraphs?
A．It will lead to widespread unemployment.
B．It may help address the labor shortage issue.
C．It will remove the need for human labor completely.
D．It doesn’t affect the nature of seasonal workers.
4．What is the author’s attitude towards the use of AI in agriculture?
A．Optimistic.
B．Distrustful.
C．Doubtful.
D．Concerned.
十、人工智能与教育
押题
词数
话题
体裁
考点
押题理由
334
一款能自主发现全新学习方
式的AI 系统，该系统在多
项测试中表现优于人类设计
的算法，打破了AI 学习规
则需由人类设定的传统模式
说明文
细节理解
推理判断
猜测词义
主旨大意
人工智能在高考阅读理解中频繁出现，文章
主要考查学生对每个环节关键内容的准确理
解，引导学生了解科技前沿、合理使用先进
技术以及对人工智能的深入思考。鲜明地彰
显了命题人旨在融入科技前沿动态的命题导
向。
In most AI’s history, machines have not truly “figured things out” on their own. They have learned — but 
only after humans first engineered the rules that allow them to learn. Even in reinforcement learning, where an AI 
agent trains itself by trial and error, the backbone of the process — the learning algorithm ( 算法) — is always 
human-designed. But a new study published in Nature hints that this division of labor may not last long. 
Researchers have created an AI system that discovered a completely new way to learn, without being told how to 
do it—and then outperformed the best human-designed algorithms on some of the hardest tests in the field. 
The researchers created a vast population of AI agents, dropped them into numerous complex environments, 
and gave them an initial learning rule. Each agent tried to solve its tasks. Their successes and failures were then 
evaluated by a separate AI system, a kind of meta-intelligence (元智能) , which acted like evolution itself. This 
meta-network did not learn the tasks. It learned how to change the learning rule itself. After each generation of 
agents, it adjusted the rule so the next generation could adapt more successfully. Over many cycles, a new learning 
rule emerged. The researchers named the discovered rule DiscoRL, and a specific implementation (实施) , tested on 
57 Atari environments, was nicknamed Disco57. 
The team used it to train a fresh AI agent and evaluated its performance against some of the strongest 
human-engineered algorithms ever developed. Disco57-trained agents outperformed all human-designed 
reinforcement algorithms. And in the hardest tests, where strategies must be invented on the fly, again, DiscoRL 
showed state-of-the-art performance, despite never being clearly programmed by humans for these domains. 
This really matters because doing well on unseen tasks is the closest thing AI has to a definition of “true 
intelligence.” It demonstrates not memorization but transferable learning. DiscoRL wasn’t just running a rule 
invented by researchers — it was using a learning strategy invented by itself.
1．Why does the author mention “the backbone of the process”?
A．To prove AI can figure out tasks independently.
B．To confirm AI’s ability to reinforce task-learning.
C．To present the fact that AI can divide backbones efficiently.
D．To illustrate that AI machines do missions programmed by humans.
2．What does Paragraph 2 mainly talk about?
A．The organizer of the study.
B．The process of the research.
C．The implication of the research.
D．The application of the study.
3．What does the underlined word “state-of-the-art” in Paragraph 3 mean?
A．Totally different.
B．Particularly disadvantaged.
C．The best available.
D．Arranged in a beautiful way.
4．What is the author’s purpose in writing the text?
A．To recommend a book on AI research. B．To present a discovery in AI development.
C．To clarify the classification of AI agents.
D．To honor an emerging AI computing company.