专题16 人工智能ChatGPT-备战高考英语阅读理解热点话题分类训练（高考模拟真题+名校最新真题）

这是一份专题16 人工智能ChatGPT-备战高考英语阅读理解热点话题分类训练（高考模拟真题+名校最新真题），共19页。试卷主要包含了推理判断题， 主旨大意题等内容，欢迎下载使用。
专题16 人工智能ChatGPT
（2023·河北唐山·统考一模）Schls in the US and elsewhere are annuncing bans n the recently released AI — pwered ChatGPT ut f fear that students culd use the technlgy t cmplete their assignments. Hwever, bans may be practically impssible given hw difficult it is t detect when text is cmpsed by ChatGPT. Is it instead time t rethink hw students are taught and evaluated?
Educatrs are starting t questin what it means t assess student learning if an AI can write an essay r paper similar t, r even better than, a student wuld — and the teacher can’t tell the difference. Many teachers believe the time-hnred learning traditin will be destryed frm the grund up by Chat GPT. The Ls Angeles Unified Schl District in Califrnia first blcked the use f ChatGPT n netwrks and devices in December 2022.
Hwever, remving technlgy frm the classrm can mean undesirable cnsequences, such as creating mre bstacles fr students with disabilities, says Trust. Additinally, restricting the use f ChatGPT n schl netwrks and devices can’t stp students frm using ChatGPT at hme and in libraries.
It is als unclear if anti-cheating sftware can reliably detect AI-assisted writing. OpenAI is wrking t develp a digital watermark that can help teachers and academics spt students wh are using ChatGPT t write essays. Open AI’s attempts t watermark AI text, hwever, hit limits.
Instead f wrrying abut hw ChatGPT culd enable cheating, educatrs shuld ask what mtivates students t cheat in the first place and wrk n develping relatinships f trust, says Jesse Stmmel at the University f Denver in Clrad.
“Talk t students really frankly abut what ChatGPT’s capable f, what it’s nt,” says Stmmel. “Have students use it t write an essay abut Jane Austen and gender dynamics, and then have them read that essay and peer review it and think abut what ChatGPT gets right and wrng.”
1．What des the authr suggest schls d?
A．Adjust teaching and assessment.
B．Meet different demands frm students.
C．Prhibit the use f ChatGPT in classrms.
D．Break with the traditinal teaching methd.
2．What is paragraph 2 mainly abut?
A．Dark future f ChatGPT.B．Educatrs’ wrrying cncern.
C．Crisis f traditinal learning.D．Difficulty in telling AI’s writing.
3．What is the authr’s attitude tward OpenAI’s watermark technlgy?
A．Amused.B．Hpeful.C．Shcked.D．Dubtful.
4．What can be inferred frm Jesse Stmmel?
A．AI helps students tell right and wrng.
B．Students shuld write abut famus writers.
C．Educatrs shuld guide students t use AI prperly.
D．The trust between teachers and students is hard t frm.
（2023·浙江杭州·统考二模）ChatGPT, designed by OpenAI t carry n cnversatins just like humans, has becme a viral excitement. The AI-pwered tl went frm zer t a millin users in just five days! Its ability t prvide in-depth answers t user questins has even drawn the attentin f distinguished technlgy cmpanies.
The intelligent rbt understands what the user says r types and then respnds in a way that makes sense. Its vast bdy f knwledge has been gathered frm the internet and archived bks. It is further trained by humans. This makes ChatGPT a useful tl fr researching almst any tpic.
“We have a lt f infrmatin n the internet, but yu nrmally have t Ggle it, then read it and then d smething with it,” says Ricard, chief science fficer and c-funder f AI cmpany Erudit. “Nw yu’ll have this resurce that can prcess the whle internet and all f the infrmatin it cntains fr yu t answer yur questin.”
ChatGPT cannt think n its wn. It depends n the infrmatin that it has been trained n. As a result, the AI tl wrks well fr things that have accurate data available. Hwever, when unsure, ChatGPT can get creative and flw ut incrrect respnses. OpenAI cautins users t check the infrmatin n matter hw lgical it sunds. Als, ChatGPT has nly been trained with infrmatin till 2021. Hence, it cannt be relied upn fr anything that happened after that.
Experts believe ChatGPT has limitless ptential t slve real-wrld prblems. It can translate lng texts int different languages, create cntent n almst any tpic, and even summarize bks.
Hwever, ChatGPT has received mixed reactins frm educatrs. Sme believe it culd serve as a valuable tl t help build literacy skills in the classrm. It culd als be used t teach students difficult science r math cncepts. But ther educatrs think ChatGPT will encurage students t cheat. They fear this will prevent them frm building critical thinking and prblem-slving skills. As a result, many districts are starting t ban its use in schls.
5．What is the unique feature f ChatGPT?
A．It has artificial intelligence.B．It can answer users’ questins.
C．It has the largest number f users.D．It can engage in meaningful cnversatins.
6．What makes ChatGPT helpful t research varius tpics?
A．Its capability f infrmatin prcessing.B．Its accurate infrmatin.
C．Its availability f up-t-date data.D．Its vast bdy f questins.
7．Why d ChatGPT users have t be cautius when using it?
A．ChatGPT is unable t think itself.B．ChatGPT lacks creativity.
C．ChatGPT ffers illgical infrmatin.D．ChatGPT is nt prperly trained.
8．What is the authr’s attitude twards ChatGPT?
A．Favrable.B．Disapprving.C．Objective.D．Intlerant.
（2023·安徽宣城·统考二模）If yu ask smething f ChatGPT, an artificial-intelligence(AI) tl, yu may immediately get the respnses, cmpletely certain and ften wrng. The questins raised by technlgies like ChatGPT cause much mre tentative answers. But they are nes that managers ught t start asking.
One issue is hw t deal with emplyees’ cncerns abut jb security. Wrries are natural. An AI that makes it easier t prcess yur expenses is ne thing; an AI that peple wuld prefer t sit next t at a dinner party quite anther. Being clear abut hw wrkers wuld redirect time and energy that is freed up by an AI helps fster acceptance. S des creating a sense f agency: research cnducted by MIT Slan Management Review and the Bstn Cnsulting Grup fund that an ability t verride an AI makes emplyees mre likely t use it.
Arthur Jag f the University f Washingtn and Glenn Carrll f the Stanfrd Graduate Schl f Business investigate hw willing peple are t give rather than earn credit—specifically fr wrk that smene did nt d n their wn. They shwed vlunteers smething attributed t a specific persn — an artwrk, say, r a business plan — and then revealed that it had been created either with the help f an algrithm r with the help f human assistants. Everyne gave less credit t prducers when they were tld they had been helped, but this effect was mre prnunced fr wrk that invlved human assistants.
The picture that emerges frm such research is messy. It is als dynamic: just as technlgies evlve, s will attitudes. But it is crystal-clear n ne thing. The impact f ChatGPT and ther AIs will depend nt just n what they can d, but als n hw they make peple feel.
9．What shuld managers take int cnsideratin when using technlgies like ChatGPT?
A．Skills used fr them.B．Issues caused by them.
C．Accuracy ensured in them.D．Alternatives discvered fr them.
10．What can we learn abut emplyees’ attitude twards AI frm Paragraph 2?
A．They find it can make their wrk mre cmplicated.
B．They think it is a threat fr AI t replace them at wrk.
C．They cnsider it’s nrmal fr AI t finish all their wrk.
D．They feel it’s vital fr them t acknwledge its abilities.
11．What did the vlunteers think f a task accmplished with AI algrithm?
A．They thught little f it.B．They were unaware f it.
C．They spke highly f it.D．They felt envius abut it.
12．What des the passage mainly talk abut?
A．The limitatin f AI.B．The imprtance f Chat GPT
C．The relatinship between AI and humans.D．The practical applicatin f Chat GPT.
（2023·广西南宁·统考一模）If yu have gne t the cinema and watched the ppular mvie “The Wandering Earth II” recently, yu might be impressed by the intelligent cmputer MOSS. It can make the reasnable decisin in the shrtest time based n the given cnditins. In fact, such strng AI like MOSS des nt exist in reality fr the time being.
Hwever, weaker AI, which replaces human prcessing in a specific field, has already influenced ur daily life. Fr example, OpenAI, the San Francisc-based startup pened up the language mdel ChatGPT fr public testing in Nvember 2022. Cmpared with traditinal search engines, which display simple results accrding t searching keywrds, ChatGPT can interact with peple just like dialguing, answer prfessinal questins, and als answer fllw-up questins. Thanks t its revlutinary user experience, ChatGPT btained mre than ne millin users just five days after its release.
Mre than a Chatbt, ChatGPT can als realize article creatin, cde generating and ther functins nly by requesting. ChatGPT can even write cde such as Pythn，debug (排错）cde and generate cmments fr cde just like a prfessinal prgrammer. It als presented strng capacity in cntent cntinuatin, literary creatin, music creatin, etc. It is s pwerful that New Yrk City public schls have banned it frm their netwrks and schl devices.
Hwever, ChatGPT is still far frm perfect. Even OpenAI's CEO nce said, “It’s a mistake t be relying n it fr anything imprtant right nw. It's a preview f prgress. We have lts f wrk t d n rbustness（稳健性）and truthfulness.”
Even thugh AI like ChatGPT still has a lng way t g, predictins abut hw the technlgy will influence ur sciety are already there. Amng the industries that may be mst affected by ChatGPT, experts say,are jurnalism, law and translatin.
13．Why is the mvie “The Wandering Earth II” mentined in the first paragraph?
A．T intrduce the tpic f the text.
B．T prve that the mvie is a hit.
C．T shw hw intelligent MOSS is.
D．T advertise the mvie.
14．Which statement abut ChatGPT will the authr prbably agree with?
A．It's well-received fr its truthfulness.
B．It can just display simple results cntent.
C．It shuld be banned frm all schl netwrks.
D．It makes human-machine interactin mre intelligent.
15．What des the OpenAI's CEO think f ChatGPT?
A．It's ttally reliable.
B．It needs imprving.
C．It's f n use.
D．It shuld speed up its prgress.
16．What is the best title fr the text?
A．ChatGPT Wrks
B．ChatGPT Sells Well
C．AI Takes Over
D．Mss Beats ChatGPT
（2023·江西赣州·统考一模）In less than 2 mnths, the artificial intelligence (AI) prgram ChatGPT has becme a cultural sensatin (轰动). ChatGPT is develped with a technique called Reinfrcement Learning frm Human Feedback t train the language mdel, enabling it t autmatically create text based n written prmpts. I asked it t rewrite the first scene f the classic American play Death f a Salesman. The utput was an amusing cnversatin. This is certainly fun, but there are serius implicatins fr generative AI prgrams like ChatGPT in science and academia.
Many cncerns relate t hw ChatGPT will change educatin. It certainly can write essays abut a range f tpics. I gave it bth an exam and a final prject. It did well finding factual answers, but the schlarly writing still has a lng way t g. If anything, the implicatins fr educatin may push academics t rethink their curses in innvative ways and give assignments that aren’t easily slved by AI. That culd be fr the best.
Mre wrrisme are the effects f ChatGPT n writing scientific papers. In a recent study abstract created by ChatGPT were submitted t academic reviewers, wh nly caught 63% f these fakes. That’s a lt f AI-generated text that culd find its way int the literature sn.
The text written by ChatGPT is nt acceptable. An AI prgram cannt be an authr. Still, accrding t Editrial Plicies, text generated by ChatGPT (r any ther AI tls) cannt be used in the wrk, nr can figures, images, r graphics be the prducts f such tls. Breaking these plicies will be cnsidered scientific miscnduct (学术不端) n different frm plagiarism (剽窃) f existing wrks.
The scientific recrd is ultimately ne f the human effrts f struggling with imprtant questins. Machines play an imprtant rle, but as tls fr the peple giving the hyptheses (假设), designing the experiments and making sense f the results. Ultimately the prduct must cme frm-and be expressed by-the wnderful cmputer in ur heads.
17．What des the underlined wrd “implicatins“ prbably mean in Paragraph1?
A．Cmpetitins.
B．Incidents.
C．Restrictins.
D．Effects.
18．What culd be ChatGPT’s benefit fr educatin, if any?
A．It culd help better the schlarly writing.
B．It culd write essays abut varieties f tpics.
C．It culd be used t design difficult assignments.
D．It culd urge academics t make imprvements.
19．What can be inferred abut ChatGPT?
A．It is welcmed by academics.
B．It may cause scientific miscnduct.
C．It is basically useless fr scientific research.
D．It wrks greatly t the advantage f educatin.
20．Which can be a suitable title fr the text?
A．ChatGPT Is Fun, But Nt an Authr
B．ChatGPT Has Becme a Cultural Sensatin
C．ChatGPT Will Change Our Way f Educatin
D．ChatGPT Plays an Imprtant Rle in Science
（2023·四川·校联考一模）Califrnia start-up OpenAI has released a chatbt (chat rbt) able t answer a variety f questins, but its impressive perfrmance has repened the debate n the risks linked t AI technlgies.
Psted n Twitter by fascinated users, the cnversatins with the chatbt called ChatGPT shw a kind f mniscient (无所不知的) machine, frm explaining scientific cncepts t writing scenes fr a play, university essays r even functinal lines f cmputer cde.
“Its answer t the questin ‘what t d if smene has a heart attack’ was incredibly clear and relevant,” Claude de Lupy, an expert in autmatic text generatin tld AFP. “When yu start asking very specific questins, ChatGPT’s respnse can be ff the mark, but its verall perfrmance remains ‘really impressive’ with a ‘high level’.” he said.
A few years ag, chatbts had the vcabulary f a dictinary and the memry f a gldfish. Nw Chatbts are getting much better at the “histry prblem” where they act in a manner cnsistent with the histry f respnses. The chatbts have graduated frm gldfish status.
Hwever, like ther prgrams relying n deep learning, ChatGPT has ne majr weakness: it desn’t have access t meaning. The sftware cannt justify its chices, such as explaining why it picked the wrds that make up its respnses.
AI technlgies able t cmmunicate are increasingly wrrying sme bservers. They vice cncern that these technlgies culd be misused t trick peple. What des ChatGPT think f the risk? “There are ptential dangers in building highly cmplex chatbts, particularly if they are designed t be indistinguishable frm humans in their language and behavir.” the chatbt said.
Open AI CEO Sam Altman shared his ideas n the debates surrunding AI. “Interesting watching peple start t debate whether pwerful AI systems shuld behave in the way users want r their creatrs intend,” he wrte, “The questin f whse values we adjust these systems t will be ne f the mst imprtant debates sciety ever has.”
21．Hw des paragraph 2 shw ChatGPT’s wnderful perfrmance?
A．By using figures.B．By listing examples.
C．By giving definitins.D．By making cmparisns.
22．What des the underlined phrase “ff the mark” in paragraph 3 prbably mean?
A．Creative.B．Inaccurate.C．Immediate.D．Objective.
23．What is the majr prblem with ChatGPT?
A．It has pr memry.B．It has limited vcabularies.
C．It fails t accunt fr its chices.D．It lacks pwer t stick t its respnses.
24．Which statement is Sam Altman mst likely t agree with?
A．The value that AI fllws remains t be discussed.
B．Advanced AI shuld g after creatrs’ intentin.
C．The functin f AI relies entirely n users.
D．Debates n the applicatin f AI are unnecessary.
（2023·四川成都·石室中学校考二模）Even if yu aren't int artificial intelligence, it's time t pay attentin t ChatGPT, because this ne is a big deal. ChatGPT is an Al chatbt system that OpenAI released in Nvember t shw ff and test what a very large, pwerful Al system can accmplish.
ChatGPT is built n tp f the OpenAI GPT-3 family f large language mdels. It lets yu type natural-language prmpts and then ffers cnversatinal respnses. It remembers the thread f yur dialgue, using previus questins and answers t infrm its next respnses. It finds relevant infrmatin in its ceans f training data, and cnverts that int plausible-sunding paragraphs f text.
Yu can ask it cuntless questins and ften will get an answer that's useful. Fr example. yu can ask it encyclpedia questins like. "Explain Newtn's laws f mtin. " Yu can tell it. “Write me a pem. " and when it des. say. "Nw make it mre exciting. ” Yu can ask it t write a cmputer prgram that'll shw yu all the different ways yu can arrange the letters f a wrd.
A few days after its launch. mre than 1 millin peple were trying ut ChatGPT. UBS analyst Llyd Walmsley estimated in February 2023 that ChatGPT reached 100 millin mnthly users in January, accmplishing in 2 mnths what tk TikTk abut 9 mnths and Instagram tw and a half years.
Many peple fear that ChatGPT might help students cheat better. But as with many ther technlgy develpments, it's nt a simple black-and-white situatin. Decades ag. students culd cpy encyclpedia entries and use calculatrs, and mre recently, they've been able t use search engines and Wikipedia. ChatGPT ffers new abilities fr everything frm helping with research t ding yur hmewrk fr yu utright.
Dustin Yrk. an assciate prfessr f cmmunicatin at Maryville University, hpes educatrs will learn t use ChatGPT as a tl and realize it can help students think critically.
" Educatrs" thught that Ggle, Wikipedia, and the Internet itself wuld ruin educatin, but they did nt. "Yrk said. "What wrries me mst are educatrs wh may actively try t discurage the acknwledgment f Al like ChatGPT. It is a tl, nt a villain. "
25．What des Paragraph 2 mainly tell us abut ChatGPT?
A．Its language mdel.
B．Its wrking thery.
C．Its design inspiratin.
D．Its develpment prcess.
26．What can't ChatGPT d?
A．Write a cmputer prgram.
B．Write hmewrk fr students.
C．Chat with users.
D．Help with husewrk.
27．Why des the authr list figures in Paragraph 4?
A．T shw the ppularity f ChatGPT.
B．T stress the high cst f inventing ChatGPT.
C．T present the creativity f artificial intelligence.
D．T prve the necessity f develping tls fr chatting nline.
28．What's the best title f this article?
A．AI advances-ur lyal servants
B．ChatGPT-merits and limits
C．A smart APP-think twice befre trusting it
D．ChatGPT-a useful AI tl
（2023春·新疆伊犁·高一校联考期中）Many educatrs fear students will use the artificial intelligence (AI) tl ChatGPT t write their reprts r cheat n hmewrk. But ther teachers are including it in the classrm.
Dnnie Piercey is a teacher in Lexingtn, Kentucky. He tld his 23 5th-grade students t try and utsmart the tl that was creating writing assignments. Piercey says his jb is t prepare students fr a wrld where knwledge f AI will be required.
He describes ChatGPT as just the newest technlgy in his 17 years f teaching that caused cncern abut the pssibility f cheating. They include tls t help with math and spelling as well as Ggle, Wikipedia and YuTube.
“As educatrs, we haven’t figured ut the best way t use artificial intelligence yet,” he added. “But it’s cming, whether we want it t r nt.”
Teaching a lwer grade level, Piercey is less wrried abut cheating than high schl teachers. His schl system has blcked students frm ChatGPT but permits teachers t use it.
Students in Piercey’s class said wrking with a chatbt makes learning fun.
Students Olivia Laksi and Katherine McCrmick, bth 10, said they can see the gd and bad parts f wrking with chatbts. They can help students wh have truble putting their thughts int writing. And there is n limit t the creativity it can add t classwrk.
McCrmick said students can use it fr suggestins, but shuld nt use it t d all the wrk.
“Yu shuldn’t take advantage f it,” McCrmick said. “Yu’re nt learning anything if yu type in what yu want, and then it gives yu the answer.”
29．What can we learn frm the first paragraph?
A．Sme teachers use the ChatGPT t teach students.
B．Many educatrs want t use ChatGPT t write their wn reprts.
C．Many students want t use ChatGPT t write their wn reprts.
D．ChatGPT can help students’ hmewrk.
30．What d we knw abut Dnnie Piercey?
A．He is a teacher f a primary schl.B．There are 23 6th-grade students in his class.
C．He has been a teacher fr 18 years.D．He has figured ut the best way t use artificial intelligence.
31．Hw des the using f ChatGPT help Students Olivia Laksi and Katherine McCrmick?
A．ChatGPT helps them cheat n their hmewrk.
B．ChatGPT leads them t play vide games well.
C．ChatGPT helps students rganize their writing skills.
D．ChatGPT limits their creativity t their classwrk.
32．What can yu get frm McCrmick’s pinin in the last paragraph?
A．Yu can learn anything frm the ChatGPT.
B．Yu can use the ChatGPT whenever yu want t .
C．Yu shuld nt use the ChatGPT t d everything fr yu .
D．Yu shuld take advantage f the ChatGPT as pssible as yu can.
参考答案：
1．A 2．B 3．D 4．C
【导语】本文是一篇说明文。主要分析了当前美国很多学校宣布禁止最近发布的人工智能驱动的ChatGPT的原因以及人们对这一问题引发的讨论。
1．细节理解题。根据首段中的“Hwever, bans may be practically impssible given hw difficult it is t detect when text is cmpsed by ChatGPT. Is it instead time t rethink hw students are taught and evaluated?(但是，鉴于检测文本何时由 ChatGPT 组成是多么困难，因此实际上不可能禁止。现在是时候重新思考如何教授和评估学生了吗？)”可知，作者认为推行禁令实际上不太可能，所以用问句的形式提议，到了该重新调整教学和评估的时候了。故选A项。
2．主旨大意题。根据第二段中的“Educatrs are starting t questin what it means t assess student learning if an AI can write an essay r paper similar t, r even better than, a student wuld — and the teacher can’t tell the difference. Many teachers believe the time-hnred learning traditin will be destryed frm the grund up by Chat GPT.(教育工作者开始质疑，如果人工智能可以写出与学生相似甚至更好的论文或论文，那么评估学生的学习意味着什么——而老师无法分辨出其中的区别。许多教师认为，历史悠久的的学习传统将被聊天GPT从头开始摧毁。)”可知，教育者开始质疑，如果老师分不清是AI还是学生写的作文时，评估学生学习情况的意义，且ChatGPT也完全摧毁这种历史悠久的学习传统，由此可知，该段主要讲的是教育者的担忧。故选B项。
3．推理判断题。根据第四段中的“OpenAI is wrking t develp a digital watermark that can help teachers and academics spt students wh are using ChatGPT t write essays. OpenAI’s attempts t watermark AI text, hwever, hit limits.(OpenAI正在努力开发一种数字水印，可以帮助教师和学者发现使用ChatGPT撰写论文的学生。然而，OpenAI为AI文本加水印的尝试达到了极限。)”可知，OpenAI正致力于开发一种数字水印，帮助教师和学者发现使用ChatGPT写论文的学生，但作者认为水印处理技术的尝试有限，由此可推断，作者对OpenAI的水印技术持怀疑态度。故选D项。
4．推理判断题。根据尾段中的““Talk t students really frankly abut what ChatGPT’s capable f, what it’s nt,” says Stmmel. “Have students use it t write an essay abut Jane Austen and gender dynamics, and then have them read that essay and peer review it and think abut what ChatGPT gets right and wrng.”(斯托梅尔说：“坦率地和学生们谈谈ChatGPT能做什么，不能做什么。”“让学生用它来写一篇关于简·奥斯汀和性别动态的文章，然后让他们阅读这篇文章，并进行同行评审，思考ChatGPT在哪些方面是对的，哪些是错的。”)”可知，Stmmel建议教育者与学生开诚布公的谈谈ChatGPT，然后让他们使用ChatGPT写一篇文章，然后对ChatGPT的文章进行评论，由此可推断Stmmel建议教育者正确引导学生使用ChatGPT。故选C项。
5．D 6．A 7．A 8．C
【导语】这是一篇说明文。文章主要介绍了人工智能ChatGPT，介绍了其特点、存在的问题以及人们对ChatGPT的看法。
5．细节理解题。根据第一段中“Its ability t prvide in-depth answers t user questins has even drawn the attentin f distinguished technlgy cmpanies.(它对用户问题提供深入解答的能力甚至引起了知名科技公司的注意)”以及第二段中“The intelligent rbt understands what the user says r types and then respnds in a way that makes sense.(这个智能机器人能理解用户所说或键入的内容，然后以合理的方式做出回应)”可知，ChatGPT的独特之处是它可以参与有意义的对话。故选D。
6．细节理解题。根据第二段“The intelligent rbt understands what the user says r types and then respnds in a way that makes sense. Its vast bdy f knwledge has been gathered frm the internet and archived bks. It is further trained by humans. This makes ChatGPT a useful tl fr researching almst any tpic.(这个智能机器人能理解用户所说或键入的内容，然后以合理的方式做出回应。它的大量知识来自互联网和存档书籍。它由人类进一步训练。这使得ChatGPT成为研究几乎任何主题的有用工具)”可知，信息处理能力使ChatGPT有助于研究各种主题。故选A。
7．细节理解题。根据第四段“ChatGPT cannt think n its wn. It depends n the infrmatin that it has been trained n. As a result, the AI tl wrks well fr things that have accurate data available. Hwever, when unsure, ChatGPT can get creative and flw ut incrrect respnses. OpenAI cautins users t check the infrmatin n matter hw lgical it sunds. Als, ChatGPT has nly been trained with infrmatin till 2021.(ChatGPT不能独立思考。这取决于它所接受的训练信息。因此，人工智能工具可以很好地处理有准确数据的事情。然而，当不确定时，ChatGPT可能会变得有创意，并给出不正确的回答。OpenAI提醒用户，无论这些信息听起来多么符合逻辑，都要检查。此外，ChatGPT的信息培训只进行到2021年)”可知，ChatGPT用户在使用时必须谨慎因为ChatGPT本身无法思考。故选A。
8．推理判断题。根据第四段“ChatGPT cannt think n its wn.(ChatGPT不能独立思考)”；第五段“Experts believe ChatGPT has limitless ptential t slve real-wrld prblems. (专家认为ChatGPT在解决现实问题方面具有无限潜力)”以及最后一段“Hwever, ChatGPT has received mixed reactins frm educatrs. (然而，教育工作者对ChatGPT的反应不一)”可推知，作者主要列举了其他人对ChatGPT的看法，没有提到自己对ChatGPT的态度，故对ChatGPT是客观态度。故选C。
9．B 10．B 11．A 12．C
【导语】这是一篇说明文。文章以人工智能ChatGPT引发的问题为引入，通过一系列研究和实验探讨了人工智能与人类之间的关系。
9．细节理解题。根据第一段“The questins raised by technlgies like ChatGPT cause much mre tentative answers. But they are nes that managers ught t start asking.(由ChatGPT等技术引起的问题带来了更多的试探性答案。但它们是经理们应该开始问的问题)”可知，管理人员在使用ChatGPT等技术时应该考虑由它们引起的问题。故选B。
10．细节理解题。根据第二段“One issue is hw t deal with emplyees’ cncerns abut jb security. Wrries are natural. An AI that makes it easier t prcess yur expenses is ne thing; an AI that peple wuld prefer t sit next t at a dinner party quite anther. (其中一个问题是如何处理员工对工作保障的担忧。担忧是自然的。让你的开支更容易处理的人工智能是一回事；一个人们在晚宴上更愿意坐在旁边的人工智能就完全是另一回事了)”可知，员工认为人工智能在工作中取代他们，是一个威胁。故选B。
11．细节理解题。根据倒数第二段“They shwed vlunteers smething attributed t a specific persn — an artwrk, say, r a business plan — and then revealed that it had been created either with the help f an algrithm r with the help f human assistants. Everyne gave less credit t prducers when they were tld they had been helped, but this effect was mre prnunced fr wrk that invlved human assistants.(他们向志愿者展示了一些属于某个特定人物的东西——比如一件艺术品或一份商业计划——然后告诉他们，这些东西是在算法的帮助下或在人工助理的帮助下创建的。当每个人都被告知他们得到了帮助时，他们对制作者的信任就会降低，但这种影响在有人工助理的工作中更加明显)”可知，志愿者不喜欢使用AI算法完成的任务。故选A。
12．主旨大意题。根据最后一段“The impact f ChatGPT and ther AIs will depend nt just n what they can d, but als n hw they make peple feel.(ChatGPT和其他人工智能的影响不仅取决于它们能做什么，还取决于它们给人的感觉)”结合文章以人工智能ChatGPT引发的问题为引入，通过一系列研究和实验探讨了人工智能与人类之间的关系。可知，这篇文章主要讲的是人工智能和人类的关系。故选C。
13．A 14．D 15．B 16．A
【导语】本文是一篇说明文。通过分析人工智能，尤其是 ChatGPT 的利与弊，表达了对其未来的期许。
【详解】推理判断题。根据第一段“If yu have gne t the cinema and watched the ppular mvie “The Wandering Earth II” recently,yu might be impressed by the intelligent cmputer MOSS. It can make the reasnable decisin in the shrtest time based n the given cnditins.In fact, such strng AI like MOSS des nt exist in reality fr the time being.” （如果你最近去电影院看了热门电影《流浪地球2》，你可能会对智能电脑MOSS印象深刻。它可以根据给定的条件在最短的时间内做出合理的决策。事实上, 像 MOSS 这样强大的 AI,现实中暂时并不存在。）和第二段首句“Hwever, weaker AI, which replaces human prcessing in a specific field, has already influenced ur daily life. ”(但是，在特定领域取代人类处理的弱人工智能已经影响了我们的日常生活。) 可知作者提及电影《流浪地球》是为了引出文章的主题“人工智能”。故答案选 A。
2.推理判断题。根据第二段倒数第二句“Cmpared with traditinal search engines, which display simple results accrding t searching keywrds, ChatGPT can interact with peple just like dialguing, answer prfessinal questins, and als answer fllw-up questins. ”（与传统搜索引擎根据搜索关键词显示简单结果相比，ChatGPT可以像对话一样与人互动，回答专业问题，也可以回答后续问题。） 可知 ChatGPT 可使人机交互更智能。故答案选 D。
3.推理判断题。由第四段中的 “We have lts f wrk t d n rbustness (稳健性) and truthfulness.” （我们在稳健性和真实性方面还有很多工作要做。”） 可知， OpenAI’s CEO 认为 ChatGPT 还需要提升。故答案选 B。
4. 主旨大意题。根据第二段的“Cmpared with traditinal search engines, which display simple results accrding t searching keywrds, ChatGPT can interact with peple just like dialguing, answer prfessinal questins, and als answer fllw-up questins.”(与传统搜索引擎根据搜索关键词显示简单结果相比，ChatGPT可以像对话一样与人互动，回答专业问题，也可以回答后续问题。),第三段的“Mre than a Chatbt, ChatGPT can als realize article creatin, cde generating and ther functins nly by requesting.”(ChatGPT不仅仅是一个聊天机器人，它还可以通过请求来实现文章创建、代码生成等功能。)以及全文内容可知，文章主要通过对比 ChatGPT 与其他智能，讲述尽管它还不完善，但是功能强大，起到很好的作用，表达了对其未来的期许，肯定了 ChatGPT 的作用和影响。因此推断A项“ChatGPT起作用”为最佳标题。故答案选 A。
17．D 18．D 19．B 20．A
【导语】本文主要介绍ChatGPT虽然可以输出文章，但它不是作者，作品必须来自我们的大脑，ChatGPT是只起着提供假设、设计实验和理解结果的工具。
17．词句猜测题。根据第二段的“Many cncerns relate t hw ChatGPT will change educatin.(许多人关注ChatGPT将如何改变教育)”可知，人们关注ChatGPT将会如何改变教育，由此猜测划线部分所在句子意为“这当然很有趣，但对于像ChatGPT这样的生成式人工智能程序在科学和学术界有着严重的影响”，即implicatins意为“影响”与affects同义。故选D。
18．细节理解题。根据第二段的“If anything, the implicatins fr educatin may push academics t rethink their curses in innvative ways and give assignments that aren’t easily slved by AI. That culd be fr the best.(如果说有什么影响的话，那就是这对教育的影响可能会促使学者们以创新的方式重新思考他们的课程，并布置一些人工智能不容易解决的作业。这可能是最好的结果)”可知，如果说ChatGPT对教育有好处的话，那就是它敦促学者们做出改进。故选D。
19．推理判断题。过呢据倒数第二段的“Still, accrding t Editrial Plicies, text generated by ChatGPT (r any ther AI tls) cannt be used in the wrk, nr can figures, images, r graphics be the prducts f such tls. Breaking these plicies will be cnsidered scientific miscnduct (学术不端) n different frm plagiarism (剽窃) f existing wrks.(尽管如此，根据编辑政策，由ChatGPT(或任何其他人工智能工具)生成的文本不能用于作品中，图形、图像或图形也不能是这些工具的产品。违反这些政策将被视为科学不端行为，与剽窃现有作品没有区别)”推知，ChatGPT可能导致科学不端行为。故选B。
20．主旨大意题。根据文章的主要内容，结合第一段的“I asked it t rewrite the first scene f the classic American play Death f a Salesman. The utput was an amusing cnversatin. (我让它重写美国经典戏剧《推销员之死》的第一幕。输出是一段有趣的谈话)”，倒数第二段的“An AI prgram cannt be an authr.(人工智能程序不能是作者)”和最后一段“The scientific recrd is ultimately ne f the human effrts f struggling with imprtant questins. Machines play an imprtant rle, but as tls fr the peple giving the hyptheses (假设), designing the experiments and making sense f the results. Ultimately the prduct must cme frm-and be expressed by-the wnderful cmputer in ur heads.(科学记录归根结底是人类努力解决重要问题的成果之一。机器扮演着重要的角色，但它是人们提供假设、设计实验和理解结果的工具。最终，产品必须来自于我们头脑中奇妙的电脑，并由它来表达)”可知，本文主要介绍ChatGPT虽然可以输出文章，但它不是作者，作品必须来自我们的大脑，ChatGPT是只起着提供假设、设计实验和理解结果的工具。因此A项“ChatGPT很有趣，但不是作者”可以作为本文的最佳标题。故选A。
21．B 22．B 23．C 24．A
【导语】本文是一篇说明文。文章主要介绍了一新款对话机器人的优异表现引起人们关于人工智能的讨论。
21．推理判断题。根据第二段中的“Psted n Twitter by fascinated users, the cnversatins with the chatbt called ChatGPT shw a kind f mniscient (无所不知的) machine, frm explaining scientific cncepts t writing scenes fr a play, university essays r even functinal lines f cmputer cde.(由着迷的用户发布在Twitter上，与名为ChatGPT的聊天机器人的对话展示了一种无所不知的机器，从解释科学概念到为戏剧，大学论文甚至计算机代码的功能行编写场景)”等多个例子可知，ChatGPT的表现令人印象深刻。由此推知，第二段是用举例的方式来展现ChatGPT的表现的。故选 B项。
22．词句猜测题。根据第三段中划线单词后的句子“but its verall perfrmance remains ‘really impressive’ with a ‘high level’.(但它的整体表现仍然“令人印象深刻”，具有“高水平”)”可知，此处表示的是“整体上ChatGPT的表现令人印象深刻”，该单词所在的句子意思与此处的内容为转折关系。 由此推知，but前的内容为“ChatGPT对具体问题的回答表现并不太好”，结合选项可知，只有inaccurate表达的意思为贬义。故选B项。
23．细节理解题。根据第五段中的“ChatGPT has ne majr weakness: it desn’t have access t meaning. The sftware cannt justify its chices, such as explaining why it picked the wrds that make up its respnses.(ChatGPT有一个主要弱点：它无法获得意义。该软件无法证明其选择的合理性，例如解释为什么它选择了构成其响应的单词)”可知，ChatGPT并不能知晓意义，无法解释自己的选择。故选C项。
24．推理判断题。根据尾段中的““Interesting watching peple start t debate whether pwerful AI systems shuld behave in the way users want r their creatrs intend,” he wrte, “The questin f whse values we adjust these systems t will be ne f the mst imprtant debates sciety ever has.”(“有趣的是，看着人们开始争论强大的人工智能系统是否应该按照用户或创造者想要的方式运行，”他写道，“我们将这些系统调整为谁的价值观的问题将是社会有史以来最重要的辩论之一。”)”可推知，Sam Altman认为关于AI追随怎样的价值观是值得社会讨论的一个话题。故选A项。
25．B 26．D 27．A 28．D
【导语】这是一篇说明文。文章介绍了ChatGPT及其作用。
25．主旨大意题。根据第二段“ChatGPT is built n tp f the OpenAI GPT-3 family f large language mdels. It lets yu type natural-language prmpts and then ffers cnversatinal respnses. It remembers the thread f yur dialgue， using previus questins and answers tinfrm its next respnses. It finds relevant infrmatin in its ceans f training data，and cnverts that int plausible-sunding paragraphs f text.( ChatGPT建立在OpenAI GPT-3家族的大型语言模型之上。它可以让你输入自然语言提示，然后提供对话应答。它会记住你对话的线索，利用之前的问题和答案来决定下一个回答。它在训练数据的海洋中找到相关信息，并将其转换为听起来合理的文本段落。)”可知，第二段是在讲ChatGPT的运作原理。故选B项。
26．细节理解题。根据第二段第二句“It lets yu type natural-language prmpts and then ffers cnversatinal respnses.(它可以让你输入自然语言提示，然后提供对话应答。)”可知，ChatGPT可以与用户聊天；根据第三段最后一句“Yu can ask it t write a cmputer prgram that’ll shw yu all the different ways yu can arrange the letters f a wrd.(你可以让它编写一个计算机程序，向你展示排列单词字母的所有不同方式。)”可 知，ChatGPT可以写程序；根据第五段最后一句“ChatGPT ffers new abilities fr everything frm helping with research t ding yur hmewrk fr yu utright.( ChatGPT提供了各种新功能，从帮助研究到完全帮你完成作业。)”可知，ChatGPT可以帮用户写作业。而文中没有提到ChatGPT可以帮助用户做家务。故选D项。
27．推理判断题。从第四段 “A few days after its launch. mre than 1 millin peple were trying ut ChatGPT. UBS analyst Llyd Walmsley estimated in February 2023 that ChatGPT reached 100 millin mnthly users in January. accmplishing in 2 mnths what tk TikTk abut 9 mnths and Instagram tw and a half years.(推出后几天。超过100万人在尝试ChatGPT。瑞银(UBS)分析师劳埃德•沃姆斯利(Llyd Walmsley)在2023年2月估计，ChatGPT在1月份的月用户数达到了1亿。TikTk用了9个月，Instagram用了2年半。)” 的数据可知，ChatGPT在极短的时间之内吸引了TikTk和Instagram需要很长时间才能吸引的用户量，所以非常受欢迎。故选A项。
28．主旨大意题。本文从第一段开始引入ChatGPT这一新事物；第二段介绍其工作原理；第三段举例介绍ChatGPT可以做什么；第四段介绍ChatGPT迅速吸引了大量用户；第五段客观分析ChatGPT其实并没有帮助学生作弊，而且帮助他们学习得更好；最后两段通过专家的观点呼吁人们客观看待ChatGPT这一工具。所以整篇文章是围绕ChatGPT来写的，应该在B、D选项中选择。本文重点介绍这一新事物作为一种工具是有用的，并没有提及其缺点。故选D项。
29．A 30．A 31．C 32．C
【导语】本文是一篇说明文。文章介绍了人工智能ChatGPT在美国学校不同受众之间的反映及看法。
29．细节理解题。根据第一段“Many educatrs fear students will use the artificial intelligence (AI) tl ChatGPT t write their reprts r cheat n hmewrk. But ther teachers are including it in the classrm.(许多教育工作者担心学生将使用人工智能工具ChatGPT写报告或在家庭作业上作弊。但其他老师也将其纳入课堂)”可知，一些老师使用ChatGPT来教学生。故选A。
30．细节理解题。根据第二段“Dnnie Piercey is a teacher in Lexingtn, Kentucky. He tld his 23 5th-grade students t try and utsmart the tl that was creating writing assignments.(Dnnie Piercey是肯塔基州列克星敦的一名教师。他告诉他的23名五年级学生，要试着智取编写写作作业的工具)”可知，Dnnie Piercey是一所小学的老师。故选A。
31．细节理解题。根据第七段“Students Olivia Laksi and Katherine McCrmick, bth 10, said they can see the gd and bad parts f wrking with chatbts. They can help students wh have truble putting their thughts int writing. And there is n limit t the creativity it can add t classwrk.( Students Olivia Laksi和Katherine McCrmick都是10岁的学生，他们说他们能看到与聊天机器人合作的好处和坏处。他们可以帮助那些在把想法写下来方面有困难的学生。它可以为课堂作业带来无限的创造力)”可知，ChatGPT帮助学生组织他们的写作技能。故选C。
32．推理判断题。根据最后一段““Yu shuldn’t take advantage f it,” McCrmick said. “Yu’re nt learning anything if yu type in what yu want, and then it gives yu the answer.”(“你不应该利用它，”McCrmick说。“如果你输入你想要的东西，然后它就会给你答案，你什么也学不到。”)”可知，McCrmick告诉我们不应该使用ChatGPT为你做所有的事情。故选C。