The artificial-intelligence chatbot ChatGPT has shaken educators since its November release. New York City public schools have banned it from their networks, and professors are improving syllabus (教学大纲) to prevent students from using it to complete homework. The chatbot’s creator, OpenAI, unveiled a tool to detect text generated by artificial intelligence to prevent abuse.
However, there is one subject area that doesn’t seem threatened. It turns out ChatGPT is quite bad at math.
While the bot gets many basic arithmetic questions correct, it makes errors when those questions are written in natural language. For example, ask ChatGPT “if a banana weighs 0.5 lbs and I have 7 lbs of bananas and nine oranges, how many pieces of fruit do I have?” The bot’s quick reply: “You have 16 pieces of fruit, seven bananas and nine oranges.”
Debarghya Das, a search-engine engineer, tried to explain why this happens in his Twitter. “Just imagine if you ask a room of people who have no idea what math is but have read many hieroglyphics (象形文字), ‘what comes after 2+2,’ they might say, ‘Usually, we see a 4.’ That’s exactly what ChatGPT is doing.” But, he adds, “math isn’t just a series of hieroplyphics, it’s computation.”
Another reason that math teachers are less worried by this revolution is that they have been here before. The field experienced dramatic changes for the first time decades ago with general availability of computers and calculators.“Math has had the biggest revolution based on the system of any mainstream subject,” said Conrad Wolfram, the strategic director of Wolfram Research, which developed Mathematica, a technical computing software program.
The broader lesson is that AI, computers and calculators aren’t simply a shortcut. Math tools require math knowledge. A calculator can’t do calculus unless you know what you’re trying to solve.
In general, AI will likely ultimately be most useful for those who already know field well: They know the questions to ask, how to identify the shortcomings and what to do with the answer.
1. What can we infer from paragraph 1 about ChatGPT?A.New York schools punished students who used it. |
B.Professors designed new syllabus with the help of it. |
C.It’s possible for teachers to identify papers written by it. |
D.The government introduced some laws to prevent its abuse. |
A.By setting an example. |
B.By drawing a comparison. |
C.By presenting some data. |
D.By telling a related story. |
A.They can use ChatGPT to solve arithmetic questions. |
B.They have already learnt the impact of technology on math. |
C.ChatGPT only understand questions written in natural language. |
D.ChatGPT usually gives misleading information in conversations. |
A.A five-year old boy who is good at calculating. |
B.A pupil who wants to cheat in the math exam. |
C.A professor who has read many hieroglyphics. |
D.A math instructor struggling with math problems. |
相似题推荐
【推荐1】San Francisco— After years of testing with a human backup(后补的) driver, General Motor’s Cruise is launching fully driverless cars onto public streets in San Francisco, the most complex urban environment robocars that have been tested, the company announced.
The company said on Wednesday it was deploying(部署) five self-driving vehicles in San Francisco’s Sunset District, an area that has many of the city’s typical traits: narrow and busy streets, steep inclines(坡度) and fog. Cruise said it was confident enough to launch the driverless trials after five years’ testing.
The company, which began the driverless testing in November, said it plans to expand to other neighborhoods. But the first tests will not put the driverless vehicle fully in control. Instead, the backup driver will be effectively moved to the passenger seat at early trials. Although those operations will not involve traditional driver controls, they will be able to stop the vehicle if there is an emergency.
Companies in Silicon Valley and elsewhere are working to make driverless vehicles a reality, saying that they will help decrease motor vehicle deaths, which number around 40, 000 annually, and allow cheaper ride-hailing(打车) trips by removing the need to pay a human driver.
Companies such as Cruise and competitor Zoox have set their sights on San Francisco, seeing the potential reward of conquering a complex urban environment and the country’s second-most populated major city, rather than starting small and gradually making progress.
1. For what reason is San Francisco chosen for the testing place?A.Its high technology. | B.Its developed economy. |
C.Its crowded roads. | D.Its complex surroundings. |
A.To operate some major functions. |
B.To decide the route of the driving. |
C.To stop the car in urgent situations. |
D.To keep the passenger seat occupied. |
A.Lower taxi fee. | B.Better urban environment. |
C.Less traffic jam. | D.Greater company profit. |
A.Benefits Brought by Driverless Cars. |
B.Driverless Cars Almost Close to Reality. |
C.Cruise Launching Driverless Cars in San Francisco. |
D.San Francisco: Perfect Testing Place for Driverless Cars. |
【推荐2】Over 70% of the earth’s surface is covered in water, but it’s undrinkable. Scientists at the University of Illinois Urbana-Champaign (UIUC) have modeled a system that can collect water vapor (水蒸汽) from above the ocean surface and freeze it into fresh drinking water at a large scale.
Water shortage is an issue that’s projected to get worse, as climate change makes dry areas even drier. The oceans, home to over 96% of water on Earth, represent a huge potential reservoir (a place to store water), but desalination (海水淡化) is a challenging and complex prospect to scale up on account of the poisonous wastewater it produces.
But nature already has a pretty good desalination system — the sun is heating up the surface of the ocean and evaporating water out of it, which of course goes on to become rain. In a new study, the UIUC team came up with a method to use this resource. The team proposes that structures could be built several kilometers offshore to obtain air rich in water vapor from above the ocean surface. That air can then be piped back to land and turned into water. This fresh water could then be used for drinking, agriculture, or whatever else a region needs it for. The whole system could be powered by offshore wind farms and land-based solar panels.
The researchers evaluated 14 cities around the world, analyzing how much water could be collected based on the offshore air in those locations. Based on their models, the scientists estimated that these structures could collect between 37.6 billion and 78.3 billion liters of water per year, depending on the conditions in a specific location.
The team says that the solution is pretty good, since it works like the natural water cycle except that the vapor is guided to where it’s needed. And while many proposed drinking water sources could become less as climate change progresses, this one should actually get even better. “The climate projections show that the oceanic vapor will only increase over time, providing even more freshwater supply,” said Afeefa Rahman, the co-author of the study. “So, the system we are proposing will be useful under climate change. This provides a much needed approach for adaptation to climate change, particularly to populations living in dry and semi-dry regions of the world.”
1. Why do scientists design this vapor-collecting system?A.To preserve water resources. | B.To relieve the water shortages. |
C.To approach climate change. | D.To reduce the pollution of water. |
A.It’s mainly powered by batteries. |
B.The water collected by it is rich in mineral. |
C.It makes full use of the natural water cycle. |
D.Structures should be built as close to the ocean as possible. |
A.The system could be applied around the world. |
B.More and more cities would introduce the system. |
C.The researchers have built many structures in 14 cities to collect water. |
D.The researchers did lots of analyses and research before introducing the system. |
A.The system can’t work in rainy regions. |
B.The system has been widely accepted by people. |
C.The system can only be used under climate change. |
D.The system will be of great help in water shortage areas. |
【推荐3】Turtles may be cute, but they are also slow and clumsy, which doesn’t really help them when they are trying to cross train tracks. Luckily, for the turtles in Japan’s Hyōgo Prefecture, railway operators and a local aquarium teamed up to find a solution.
Between 2002 and 2014, disruptions (中断) of train operation caused by turtles were reported 13 times, with many more probably going unreported. That is why, in 2015, West Japan Railway Co. and Suma Aqualife Park in Kobe joint forces to prevent tragic turtle deaths on the tracks and unnecessary train delay. After running a series of tests and experiments, they came up with a U-shaped ditch (沟渠 ) that collects the turtles, preventing them from getting stuck between the metal tracks and getting crushed to death.
Tests showed that turtles trying to cross the train tracks often fall into the space between them, which leaves them with no other option than to walk between them. Sadly, this causes some of them to get stuck between the metal tracks at junctions(交叉口), and they unavoidably get crushed to death when the points blades(叶片)move.
“The turtles are basically just going about their daily business and have to cross the lines to get to a pond,” a railway spokesperson said. “When the points blades move, unfortunately they get squashed between them and die. They can cause long delays to operations so we consulted with a turtle specialist to find the best way to help them.”
The U -shaped concrete ditches installed at several critical points along the train lines near the city of Kobe catch the turtles at danger of getting stuck between the tracks. They allow the turtles to get past the junctions safely. 10 turtles were reportedly saved in the first month since the U-shaped ditch was put to use in November 2015.
1. What’s the best title of the text?A.An Important Invention in Japan |
B.The U-shaped Concrete Ditches |
C.Turtles Crossing Train Tracks in Danger |
D.Japan’s Clever Solution to Help Turtles Cross Train Tracks |
A.Why the U-shaped ditch is invented. |
B.The turtles’ terrible conditions. |
C.The train delay caused by turtles. |
D.How the turtles go through the tracks. |
A.The turtles. | B.The trains. |
C.The train tracks. | D.The points blades. |
A.They are installed on the railway tracks. |
B.They can prevent turtles from escaping. |
C.They are easy to trap the turtles. |
D.They are helpful for turtles to cross train tracks. |
【推荐1】Kite surfing is a sport that takes place on the surface of the water, on top of boards that are similar in design to wakeboards (尾波滑水板), with the surfer pulled by a kite. The popularity of kite surfing has grown, although it is rather difficult to master. Kite surfing uses the same equipment as kite boarding, but is generally given more to riding on top of waves, rather than on smaller lakes or other calm bodies of water.
Early experiments in kite transportation took place through the 19th century, and at the dawn of the 20th century Samuel Cody crossed the English Channel using kites and a small boat. Kite surfing itself, however, wasn’t really born until the late 1970s. Beginning in that decade, more people became involved in using kites for personal transport and sport. Kite technology itself advanced enormously during the 1970s, with kites becoming much more controllable.
The early kite surfing was rather accidental, but innovators continued to develop the technology to make it easier to control and safer. By the late 1990s, kite surfing was becoming popular off the coast of Maui, France and throughout the world. In 1997, a special kite system was built specifically for kite surfing, which helped kite surfing into the mainstream.
Because of the high speeds and the dangers, it is generally recommended that would-be kite surfers take at least a few basic classes to learn the fundamentals. The main attraction of kite surfing is the potential for amazing tricks. The kite allows the surfer to experience enormous jumps, flying through the air for large distances before hitting the water in a controlled fashion. Like skateboarding, kite surfing makes use of a number of board grabs, with complex tricks possible due to the long period of time the surfer can remain in the air.
1. What does the author mainly intend to tell us in Paragraph 1?A.What is kite surfing. | B.Where to go kite surfing. |
C.How to enjoy kite surfing. | D.Why people like kite surfing. |
A.sport | B.experiment |
C.entertainment | D.transportation |
A.Kite surfing often leads to accidents. |
B.Maui is home to kite surfing. |
C.Kite control is the key to kite surfing. |
D.Kite surfing came into being in the 1990s. |
A.Its potential for tricks. | B.Its long time in the air. |
C.Its speed across the water. | D.Its big distance over the water. |
【推荐2】It happened to me recently. I was telling someone how much I had enjoyed reading Barack Obama’s Dreams From My Father and how it had changed my views of our President. A friend I was talking to agreed with me that it was, in his words, “a brilliantly written book”. However, he then went on to talk about Mr. Obama in a way which suggested he had no idea of his background at ail. I sensed that I was talking to a book liar.
And it seems that my friend is not the only one. Approximately two thirds of people have lied about reading a book which they haven’t. In the World Book Day’s “Report on Guilty Secrets”, Dreams From My Father is at number 9. The report lists ten books, and various authors, which people have lied about reading, and as I’m not one to lie too often (I’d hate to be caught out), I will admit here and now that I haven’t read the entire top ten. But I’m pleased to say that, unlike 42 percent of people, I have read the book at number one, Gorge Orwell’s 1984. I think it’s really brilliant.
The World Book Day report also has some other interesting information in it. It says that many people lie about having read Jane Austin Austen, Charles Dickens, Fyodor Dostoevsky (I haven't read him, but haven’t lied about it either) and Herman Melville.
Asked why they lied, the most common reason was to “impress” someone they are speaking to. This could be tricky if the conversation became more in-depth!
But when asked which authors they actually enjoy, people named J.K. Rowling, John Grisham, Sophie Kinsella (ah, the big sellers, in other words). Forty-two percent of people asked admitted they turned to the back of the book to read the end before finishing this story (I will come clean: I do this and am astonished that 58 percent said they had never done so.).
1. How did the author find his friend a book liar?A.By judging his manner of speaking. |
B.By looking into his background. |
C.By mentioning a famous name. |
D.By discussing the book itself. |
A.Charles Dickens is very low on the top-ten list. |
B.42% of people pretended to have read 1984. |
C.The author admitted having read 9 books. |
D.Dreams From My Father is hardly read. |
A.control the conversation |
B.appear knowledgeable |
C.learn about the book |
D.make more friends |
A.Favorable. | B.Uncaring. |
C.Doubtful. | D.Friendly. |
【推荐3】Lake Titicaca is a freshwater lake in the Andes mountains on the border of Bolivia. It’s the largest lake in South America. Dotted along the lake’s western corner, you can find dozens of artificial islands. Several thatched (茅草的) houses and structures sit on each island, some of which are only 30 meters or so wide, although larger ones exist.
The floating platforms are built by piling layers upon layers of mixed totora roots and reeds (芦苇). The totora plant is necessary to life on the lake. Its strong root fibers are used to make houses, boats, roofs, mattresses, and more, as well as used to make tea, traditional medicine, and tasty dishes known as “lake banana”.
So the story goes, this unusual living arrangement came out in the pre-Columbian era when the ambitious Inc a Empire began invading (入侵) into the mainland villages of Uros. The Uros villages moved to Lake Titicaca where they built these floating platforms. If invading Inca came their way, the Uros could push these artificial islands out to the middle of the lake and flee their attack.
The threat of the Inca Empire has long passed, but the tradition remains strong. Even today, it’s estimated that around 1,300 Uros live on some 100 constructed islands on Lake Titicaca. Maintaining the lifestyle isn’t easy. While totora is a tough material, the thatched organic material degrades due to the forces of nature over time, meaning the islands and homes are in a constant state of repair.
However, the traditional way of life isn’t without its modern benefits. Many of the islands have solar panels that power lights, radios, and television satellites. The lake’s artificial islands have also become a must-see spectacle for tourists, with some locals even renting out their totora-thatched homes to travelers on online rental website Airbnb.
1. What is the second paragraph mainly about?A.The value of the totora plant. | B.The process to build the islands. |
C.The islands made of water plants. | D.The tradition and history of the Uros. |
A.To live a nature-friendly life. | B.To preserve their old lifestyle. |
C.To build thatched houses on them. | D.To protect themselves against enemies. |
A.They have lost their native language. |
B.Modern devices have come into their life. |
C.Tourism is their major source of income. |
D.They’re tired of repairing the totora-thatched houses. |
A.To persuade. | B.To educate. | C.To advertise. | D.To introduce. |
【推荐1】Microsoft says it has used the natural language AI ChatGPT to control robots with simple text commands. The approach means people with no engineering experience will be able to instruct robots to carry out tasks.
Microsoft said the research was intended “to see if ChatGPT can think beyond text, and reason about the physical world to help with robotics tasks”. Robots are typically controlled by software that has been written by humans and sets out precise instructions, or else by some form of network AI that can learn to carry out tasks based on large numbers of examples.
Neither Microsoft nor Open AI responded to a request for comment on this, but Microsoft said ChatGPT allows a user to monitor the process. ChatGPT isn’t in direct, real-time control of the robot, but simply creates the code that controls it. Microsoft said ChatGPT output shouldn’t be used directly to control a robot without “careful analysis”, but experts warn that it is risky to even begin thinking about placing AI in control of physical machines.
Mark Coeckelbergh at the University of Vienna, Austria, says that regulations may be needed in order to set out where AI can be used and who is responsible should things go wrong. “It’s a very dangerous thing to just say ‘let’s give control of these computers to AI’,” says Coeckelbergh. “The problem with contemporary artificial intelligence is that it’s not transparent to the user. That’s a huge problem.”
Others were doubtful about the ability of language models to program robots for complex tasks. Kathleen Richardson at De Montfort University Leicester, UK, says that AI language models are convincing mimics (模仿者), but that their real ability — including their adaptability to control robots — is often overblown. “I think Microsoft, and most people who design robots, overstate what they can and can’t do,” she says.
1. What’s the purpose of the research conducted by Microsoft?A.To prove ChatGPT’s capacity to carry out tasks. | B.To test ChatGPT’s ability to control robots. |
C.To teach ChatGPT to give accurate instructions. | D.To identify the hidden problems of ChatGPT. |
A.Curious. | B.indifferent. | C.Supportive. | D.Opposed. |
A.It’s safe to use AI to control physical machines. | B.AI’s adaptability to control robots is reliable. |
C.Regulations of AI use should be in place. | D.The problem with AI is still under debate. |
A.ChatGPT tells robots what to do. | B.Power of ChatGPT is changing us. |
C.Problems of ChatGPT need solving. | D.AI takes control of robots wisely. |
【推荐2】Will chatbots that can generate fascinating articles destroy education as we know it?
New York City’s Department of Education recently banned (禁止) the use of ChatGPT. “While the tool may be able to provide quick and easy answers to questions,” says the official statement, “it does not build critical-thinking skills, which are necessary for academic and lifelong success.”
Banning such use of technology from the classroom is a nearsighted response. Instead, we must find a way forward in which such technologies combine well with, rather than replace, student thinking.
Banning ChatGPT is impossible in practice. Students will find ways around the ban, which will cause a further defensive response from teachers and administrators, and so on. It’s hard to believe that a close race between those digital natives and their educators will end in a decisive victory for the latter. In fact, chatbots may well speed up a trend (趋向) toward valuing critical thinking. In a world where computers can fluently answer any question, students need to get much better at deciding what questions to ask and how to fact-check the answers the program generates.
So how do we encourage young people to use their minds when real thinking is so hard to tell apart from its simulacrum (假象)? Teachers, of course, will still want to watch students taking old-fashioned, in-person, no-chatbot-allowed exams to check that they do not cheat.
But we must also figure out how to do something new: how to use tools like GPT to inspire deeper thinking. GPT often generates text that is fluent and “reasonable” — but wrong. So using it requires the same mental heavy lifting that writing does: forming an opinion, creating an outline, picking which points to explain and which to drop, and looking for supporting facts. GPT can help with those tasks, but it can’t put them all together. Writing a good essay still requires lots of human thought and work. Indeed, writing is thinking, and good writing is good thinking.
One approach is to focus on the process as much as the result. For instance, teachers might require four drafts of an essay. After all, as John McPhee, the famous writer, said, “the central nature of the process is revision.” Each draft gets feedback from the teacher, from peers or even from a chatbot. Then the students produce the next draft, and so on.
Will AI one day outperform human beings in thinking? Maybe, but for now, we must think for ourselves. Like any tool, GPT is an enemy of thinking only if we fail to find ways to make it our partner.
1. How does the author feel about the ban?A.Understandable. | B.Irresponsible. | C.Unwise. | D.Necessary. |
A.Because students are digitally fluent. | B.Because schools will defend the ban. |
C.Because ChatGPT will keep developing. | D.Because people treasure critical thinking. |
A.By quoting others. | B.By presenting facts. | C.By giving examples. | D.By showing similarities. |
A.Is GPT a process or a result? | B.Will GPT outperform students? |
C.Why Chatbots become a new trend? | D.How can Chatbots serve education? |
【推荐3】As Geoffrey Hinton, widely seen as the godfather of artificial intelligence, warns about growing dangers from how it is developing, businesses are scrambling to see how they can use the technology to their advantage.
Recently, the hot topic is how to employ ChatGPT style technology across their businesses as quickly as possible. A boss looked at his computer, entered the transcript of a customer complaint call, and asked ChatGPT to respond to it based on set of rules. In about a minute it came up with a very credible answer, which was 85% accurate but only cost 1% of the cost of employing staff. The good news for all is the pure enhancement to productivity.
Technology has improved the quality of life. A lot of innovation has made our entertainment time more enjoyable. But the real shock has been that these technologies are usable in a commercial context, not just for “low-cognitive, repetitive” tasks, i. e. robotic-tasks likely to be automated.
The surprise has been how employable the technology is to highly creative, high-value work. Specifically, creating an image, or music, or coding a programme can be achieved in seconds rather than weeks of briefing and refining. So the good news is that rapid employment of this technology could solve the longstanding productivity crisis. The bad news is that it could occur so rapidly as to overtake workers’ ability to adapt in time, creating social crisis. Could we face in call centers and creative studios in the 2020s, the equivalent of what happened in the coal mines in the 1980s?
Some individuals have started to suggest that states should provide a basic income due to the possibility of AI leading to job loss. The response of tech enthusiasts is that “You won’t be replaced by an AI, but you might be replaced by someone who knows how to use AI”. But they used to say that’s why everybody should learn how to code. That might not be such wise career advice any more.
1. What do we learn about ChatGPT?A.It improves customers’ satisfaction. |
B.It enhances productivity and reduces cost. |
C.It perfects customers’ ability to summarize. |
D.It revolutionizes the business with accuracy. |
A.By enriching leisure time. | B.By increasing dependence. |
C.By simplifying people’s work. | D.By reshaping entertainment forms. |
A.Alternative. | B.Opposite. | C.Similarity. | D.Reference. |
A.Intolerant. | B.Favorable. | C.Indifferent. | D.Objective. |