第1部分 训练4(Word教参)-【精讲精练】2026年高考英语二轮专题辅导与训练

训练4

[对应学生用书P206]

Ⅰ.阅读理解

C

(2025·长沙新高考适应性考试)Artificial intelligence—or AI—is the computer code (代码) that allows a machine to do something that normally requires a human brain.On social media，for instance，AI sorts the posts so that the first ones you see are likely to be those you'd prefer.But before AI can do any of that，developers must train it.And the training consumes energy.

AI training is powered mostly by fossil(化石) fuels，according to one of the largest studies of its kind.Less than 25 percent of AI uses low-carbon energy sources such as hydroelectricity and nuclear power during training.In fact，researchers now worry that training's appetite for energy could soon become a huge problem.The energy to develop AI comes out of the electrical net.And in most parts of the world，making electricity emits (排放) carbon dioxide and other greenhouse gases into the air.

Language-processing AI systems learn by devouring (吞食) texts in a particular language.This might include “reading” everything ever published online in some language，including libraries full of old books that have been digitalized.Such data-intensive training uses a lot of energy.To translate between English and Chinese，for example，an AI model may deal with millions or even billions of translated books and articles.In this way，it learns which words and phrases match.Later，when given a new text，it recommends its own translation.

In 2019，researchers calculated the impact of developing the AI model named Transformer.It released a massive 626，000 pounds of carbon dioxide.That's equal to greenhouse gases that would be poured by five American cars from when they were made to when they were junked.

Today，most AI development happens at data centers.These computer-filled buildings account for some 2 percent of US electricity use and 1 percent of global energy use.Some AI experts have sounded an alarm about the threat these energy giants pose.

[语篇解读]　本文是一篇说明文。人工智能给人们的生活带来了很多便利，但是开发人工智能需要消耗大量能源，会向空气中排放二氧化碳和其他温室气体。

28．What is AI training mainly energized by?

A．Fossil fuels.    B．Solar energy.

C．Nuclear energy.    D．Greenhouse gases.

解析　细节理解题。根据第二段中的“AI training is powered mostly by fossil (化石) fuels”可知，人工智能训练主要由化石燃料提供动力。原文的“fossil fuels”与A项为原词复现。 故选A。

答案　A

29．What do we know about the language-processing AI systems?

A．They process limited texts.

B．They will replace translators.

C．They seldom read digital books.

D．They are energy-consuming.

解析　细节理解题。根据第三段中的“Language-processing AI systems learn...a lot of energy”可知，语言处理人工智能系统通过吞食特定语言的文本来学习。这可能包括“阅读”网上用某种语言出版的所有东西，包括涵盖了已数字化旧书的系列丛书。这种数据密集型训练需要消耗大量能源。文中的“uses a lot of energy”与D项中的energy-consuming 是同义转换。故选D。

答案　D

30．Why does the author mention Transformer in paragraph 4?

A．To highlight its significance.

B．To speak highly of AI's power.

C．To demonstrate researchers' creativity.

D．To stress its effect on the environment.

解析　写作意图题。根据第四段中的“In 2019，researchers calculated the impact of developing...when they were made to when they were junked”可知，2019年，研究人员推测了开发名为Transformer的人工智能模型的影响。它释放了626 000磅的一氧化碳，这相当于五辆美国汽车从制造到报废所排放的温室气体。提到Transformer是为了强调它对于环境的影响。故选D。

答案　D

31．What does the author imply in the last paragraph?

A．AI's development poses a threat to data centers.

B．The energy consumption of AI can't be ignored.

C．AI is effectively transforming the way we learn.

D．People should take action to reduce gas emissions.

解析　推理判断题。根据最后一段可知，大多数人工智能开发都在数据中心进行。这些布满计算机的建筑占美国用电量的2%，占全球用电量的1%。一些人工智能专家对这些能源巨头构成的威胁发出了警告。从中我们可以推断出，人工智能的能耗是不容忽视的，故选B。

答案　B

[二次精读]

D

(2025·南京六校联合调研)One day，my 11-year-old daughter told me that she wanted to return a pair of pants on an online shopping platform.“You need to keep clicking the ‘something else’ button if you want to talk to a real person，” I told her.She followed my word，and after a bit of clicking，she managed to talk to a real person and get the pants returned.In the process，I had the chance to think about what this everyday interaction was teaching my daughter.

I thought about what it meant to constantly ask for “something else”．It's not only about having a real person to talk with.It's equally about the quality of our communication.It's about thinking something，then saying it，and then having another person think and say something back to you.It seems simple and yet it may become the exception，not the rule.In many cases，it already has.

With the “quick responses” at the bottom of the mailbox，potential responses often appear right in front of us.“Yes，sure.”“Of course！” or “I got it.” The responses，I've noticed，are mostly okay.Sometimes，they're exactly what I want to say.Occasionally，they're wildly wrong.At first，I told myself I wouldn't use those pre-made responses.Soon enough，though，I got in a rush and found they were good.

What would happen if our language got automated in general in this way? What if，instead of generating responses，we simply pointed and clicked? Choosing a response is a kind of thinking，but it's far less accurate than the kind needed to generate content.It has to do more with recognizing a “good enough” response.

Instead of starting with a blank board，we critique (评判) things.That's not useless，but it