Лабораторный журнал. Запись 1. Урок аудирования

Идея

Нам нужно выбрать тему и материал. В статье², мы раскрывали упражнения, посвященные развитию навыка чтения. Выбранная нами тема была посвящена “Технологический суверенитет страны”. Здесь, на страницах этого лабораторного журнала, мы пойдем немного иным путем и попробуем посмотреть на изменение этики с развитием технологий ИИ. Развивать будем не навык чтения, а навык аудирования. Не уверен, что тема останется именно той, которую задумали. Возможно что-то измениться, но не сильно.

Подбор материала

Поиграемся немного с поисковыми запросами. Появившиеся инструменты ИИ вынуждают обращаться к ним, а не гуглу, не потому что он хуже, а потому что ответы гугла, без контекста. Вместо него обращаемся к perplexity.ai. Стоит отметить, что в проектировании этого урока используются несколько инструментов ИИ. Наш запрос к ИИ написан так:

I am looking for videos on YouTube that cover the topic of the new ethics of the Internet in the age of all-pervasive AI technologies. I’m interested in understanding the caveats and benefits AI brings and how ethics have changed because of this. I need short and concise videos, around 3-5 minutes in length, that would be suitable for use in ESL classes^1.

Система предлагает три видео и небольшое саммари каждого из них:

This video features Pascale Fung, director of the Centre for AI Research, who talks about the importance of ethics in AI. She discusses the potential benefits of AI, as well as the potential risks and challenges that come with it. This video is 3 minutes and 24 seconds long.

In this video, Peter Norvig, Director of Research at Google, discusses the ethical considerations surrounding AI. He talks about the potential benefits of AI, as well as the potential risks and challenges that come with it. This video is 4 minutes and 57 seconds long.

This video features Leon Furze, an expert in AI in education, who discusses the use of AI in the classroom. He talks about the potential benefits of AI, as well as the potential risks and challenges that come with it. This video is 5 minutes and 54 seconds long.

Все видео достаточно долгие, около часа, скорее всего алгоритм указывает на какой-то один фрагмент. Анализирую оглавления имеющееся у роликов, удалось найти парочку интересных разделов, но для обучения класса, слишком не структурная информация. Это будет сложно, нужно исправлять логику повествования, вводить специфическую, разговорную лексику. В одном из видео разговор ведется с австралийцем, это тоже требуется некоторой подготовки. В общем, нужно что-то иное. Немного скорректируем запрос, вместо video, используем TED education video. Сразу получаем джекпот.

Любопытный материал вышел три месяца назад.
Why AI is incredibly smart and Shockingly stupid from Yejin Choi

Скорее всего английский не ее родной язык, но на сцену TED без серьезной подготовки не выпускают. Посмотрим.

Видео, хронометражам 16 минут из которых 13 ее речь. Много технических терминов и интересных философских вопросов. Возьмем его и попробуем построить вокруг него урок аудирования.

[!NOTE] Personal note
Скорее всего есть более интересные видео, но мы не будем тратить время на более детальный подбор, вместо этого построим короткий 45 минутный урок.

Проектируем урок

После каждого урока ученик должен становиться лучше в конце чем он был в начале. Для решения этой задачи необходимо подобрать интересный культурный аспект и в этом видео он как раз имеется.

[!NOTE] Cultural angle
Common sense is not so common. Voltaire

Попробуем раскрыть тему здравого смысла и в чем он выражается в современном мире. Придумаем название уроку. Чем провокационней, тем интересней, используем часть названия из оригинального TED выступления, но оставим многое за рамками, чтобы было интересно с самого начала. У детей не будет доступа к остальному материалу, так как он будет постепенно им предъявляться.

[!NOTE] Title
Incredibly smart and shockingly stupid

Зададим тон урока, через discussion topics, три простых вопроса. Мы, конечно, можем их придумать самостоятельно, но нам же интересно попробовать в этом деле ИИ, так и сделаем. Обратимся к twee.com. Переходим в раздел speaking, далее выбираем find discussion questions, мы используем уровень advanced. Зададим несколько тем, и подберем три наиболее подходящих вопроса для разогрева.

[!NOTE] Первая тема
Запрос: Common sense is not so common, Voltaire
Ответ Twee.com:

1. Who is Voltaire and what was his contribution to literature?
2. What does the quote «Common sense is not so common» mean to you?
3. Can you think of any examples where people lack common sense in everyday life?
4. Do you believe that common sense can be taught or is it something that comes naturally?
5. How important do you think common sense is in making decisions?
6. In your opinion, what are some factors that contribute to a person’s level of common sense?
7. Have you ever been in a situation where someone lacked common sense? How did you handle it?
8. Is there a difference between intelligence and common sense? If so, how would you define each?
9. Why do you think some people have more common sense than others?
10. Can you think of any ways to improve one’s common sense?

[!NOTE] Вторая тема
Запрос: Incredibly smart and shockingly stupid computer
Ответ Twee.com:

1. What is the most impressive feat of artificial intelligence that you have heard of?
2. Can you think of any examples where a computer has made a mistake or error in judgment?
3. How do you think advancements in AI will impact our daily lives in the future?
4. Have you ever interacted with a chatbot or virtual assistant? If so, what was your experience like?
5. Do you believe that computers can truly be considered «intelligent»? Why or why not?
6. Are there any ethical concerns surrounding the development and use of advanced AI technology?
7. In your opinion, what are some potential benefits and drawbacks to relying heavily on computers for decision-making processes?
8. Can you think of any industries or fields that could greatly benefit from the implementation of AI technology?
9. How do you think humans can ensure that they remain in control of AI systems, rather than the other way around?
10. Do you think it’s possible for a computer to become too intelligent for its own good? Why or why not?

Курсивом выделены вопросы, которые интересно было бы обсудить, и начать разогрев урока.

The beginning

Exercise №1 (⏳5 min to complete. 5 out of 45)

You are going to watch a TED talk by Yejin Choi Why AI Is Incredibly Smart and Shockingly Stupid, read the text about the speaker and the talk. Then work in pairs or small groups and discuss the questions

1. Can you think of any examples where people lack common sense in everyday life?
2. What is the most impressive feat of artificial intelligence that you have heard of?
3. Have you ever interacted with a chatbot or virtual assistant? If so, what was your experience like?

В качестве информации о спикере и выступлении возьмем инфо с YouTube страницы выступления и страницы википедии о спикере.

BIO. The speaker and the speech

[!NOTE] Short BIO
Yejin Choi is the Brett Helsel Professor of Computer Science at the University of Washington. Her research considers natural language processing and computer vision. She is on the TED stage to demystify the current state of massive artificial intelligence systems like ChatGPT, highlighting three key problems with cutting-edge large language models (including some funny instances of them failing at basic commonsense reasoning.) She welcomes us into a new era in which AI is becoming almost like a new intellectual species – and identifies the benefits of building smaller AI systems trained on human norms and values.

Следующим этапом, нам необходимо ввести лексику из видео. Для этого необходимо получить текстовую версию фрагмента, который планируется к прослушиванию. С 0:00 до 2:57, вводная часть предлагает достаточное количество новых слов. Выделены курсивом. Для этой работы используем инструмент YouTube Video to Text на Twee.com.

[!NOTE] Transcript №1
So, I’m excited to share a few spicy thoughts on artificial intelligence. But first, let’s get philosophical. By starting with this quote by Voltaire, an 18th century Enlightenment philosopher who said, Common sense is not so common.

Turns out, this quote couldn’t be more relevant to artificial intelligence today. Despite that, AI is an undeniably powerful tool, beating the world-class Go champion, acing college admission tests, and even passing the bar exam.

I’m a computer scientist of 20 years, and I work on artificial intelligence. I am here to demystify AI.

So AI today is like a Goliath. It is literally very, very large. It is speculated that the recent ones are trained on tens of thousands of GPUs and a trillion of words. Such extreme-scale AI models, often referred to as large-language models, appears to demonstrate sparks of AGI, (artificial general intelligence), except when it makes small, silly mistakes, which it often does. Many believe that whatever mistakes AI makes today can be easily fixed with brute-force bigger scale and more resources. What possibly could go wrong?

So there are three immediate challenges we face already at the societal level. First, extreme-scale AI models are so expensive to train, and only a few tech companies can afford to do so. So we already see the concentration of power. But what’s worse for AI safety? We’re now at the mercy of those few tech companies, because researchers in the larger community do not have the means to truly inspect and dissect these models. And let’s not forget their massive carbon footprint and the environmental impact.

And then there are these additional intellectual questions. Can AI, without robust common sense, be truly safe for humanity? And is brute-force scale really safe?

Для выделенных слов мы попросим Twee.com составить предложения, инструмент Sentences with target vocabulary.

Essential Vocabulary

Exercise №2 (⏳5 min to complete. 10 out of 45)

Read the sentences (1-8). Words in bold are used in the TED talk. First guess the meaning of the words and collocations. Then match their definitions (a-h), and use at least three of them in sentences of your own.

[!NOTE] Target sentences

1. My mind is constantly filled with spicy thoughts about the intricacies of quantum mechanics and parallel universes.
2. Through conducting experiments in the lab, we were able to demystify the process of gene editing for our study.
3. As a group, we often speculate on the potential long-term effects of climate change on global ecosystems.
4. The sparks of creativity ignited within me as I gazed out at the vibrant sunset over the ocean.
5. In order to crack the sophisticated encryption code, we had to utilize brute-force methods to manually test every possible combination.
6. Travelling alone in an unfamiliar city, I felt at the mercy of the bustling crowds around me.
7. Using advanced imaging techniques, we were able to meticulously dissect the intricate anatomy of the insect under our microscope.
8. Our research team conducted multiple trials to ensure that our experimental model was robust enough to withstand changing environmental conditions.

Попросим подобрать определения к каждому из слов и выражений, используем инструмент Word definition matching:

[!NOTE] Definitions
a. To make something easier to understand by explaining it in simpler terms or breaking down complex ideas.
b. To carefully examine and analyze something by breaking it down into its individual parts or components.
c. Small bursts or flashes of something, such as creativity, inspiration or passion.
d. Thoughts that are exciting, provocative or controversial in nature.
e. Being completely vulnerable and dependent on someone else’s actions or decisions.
f. Strong, healthy and able to withstand difficult conditions or challenges.
g. To form an opinion or guess about something without having all the facts or evidence.
h. A method of solving a problem through sheer force or strength rather than using finesse or strategy.

После этого мы попросим прослушать учащихся первую часть видео с 0:00 до 2:57 и ответить на вопросы, которые Twee.com любезно предоставит. Курсивом выделены вопросы которые останутся в итоговом упражнении. Для этого мы используем инструмент Create question for a YouTube video.

Watching video

Exercise №3 (⏳7 min to complete. 17 out of 45)

Watch the first part (⏳00:00 – 2:57 min) twice, then work in pairs os small groups to answer the questions. (⏳7 мин на выполнение задания в классе)

[!NOTE] Create questions for a YouTube Video
Questions:
> 1. What is the relevance of Voltaire’s quote to artificial intelligence?

2. What are some examples of AI’s capabilities?
3. What is the speaker’s profession and how long have they worked on AI?
4. What is the current state of AI?
5. What are large language models and what do they demonstrate?
6. What are some of the mistakes that AI makes?
7. How do many people believe that AI’s mistakes can be fixed?
8. What are the three immediate challenges that society faces with extreme-scale AI models?
9. Why are researchers in the larger community unable to inspect and dissect these models?
10. What are some intellectual questions surrounding AI’s safety and scalability?

Correct answers:
> 1. The quote is relevant because AI is powerful but lacks common sense.

2. AI can beat world-class champions, ace college admission tests, and pass the bar exam.
3. The speaker is a computer scientist of 20 years who works on AI.
4. AI is like a Goliath, very large and powerful.
5. Large language models are trained on tens of thousands of GPUs and a trillion words, and they demonstrate sparks of AGI.
6. AI often makes small, silly mistakes.
7. Many people believe that AI’s mistakes can be fixed with brute force, bigger scale, and more resources.
8. The three immediate challenges are the concentration of power among a few tech companies, the inability of researchers to inspect and dissect these models, and the environmental impact of their massive carbon footprint.
9. Researchers in the larger community do not have the means to inspect and dissect these models because they are so expensive to train.
10. Intellectual questions include whether AI without robust common sense can be truly safe for humanity and whether brute-force scale is really the only way.

Следующим шагом, мы представим учащимся еще несколько словарных значений, однако на этот раз им придется заполнять пропуски в соответствующих предложениях. Для этого нам снова придется создать транскрипт следующей части видео (⏳02:57 – 5:48 min). Предложения курсивым будут использованы в упражнении, а выделенные слова являются предметом поиска.

[!NOTE] Transcript №2
We need to make AI smaller to democratize it, and we need to make AI safer by teaching human norms and values.

Perhaps we can draw an analogy from David and Goliath. Here, Goliath being the extreme-scale language models, and seek inspiration from an old-time classic, The Art of War, which tells us, in my interpretation, know your enemy, choose your battles, and innovate your weapons.

Let’s start with the first, know your enemy, which means we need to evaluate AI with scrutiny. AI is passing the bar exam. Does that mean that AI is robust at common sense? You might assume so, but you never know.

So suppose I left five clothes to dry out in the sun, and it took them five hours to dry completely. How long would it take to dry 30 clothes? GPT-4, the newest, greatest AI system, says 30 hours. Not good.

A different one. I have a 12-liter jug and a 6-liter jug, and I want to measure 6 liters. How do I do it? Just use the 6-liter jug, right? GPT-4 spits out some very elaborate nonsense. Step one, fill the 6-liter jug. Step two, pour the water from 6- to 12-liter jug. Step three, fill the 6-liter jug again. Step four, very carefully pour the water from 6- to 12-liter jug. And finally, you have 6 liters of water in the 6-liter jug that should be empty by now.

OK, one more. Would I get a flat tire by bicycling over a bridge that is suspended over nails, screws and broken glass? Yes, highly likely, GPT-4 says. Presumably because it cannot correctly reason that if a bridge is suspended over the broken nails and broken glass, then the surface of the bridge doesn’t touch these sharp objects directly.

OK, so how would you feel about an AI lawyer that aced the bar exam yet randomly fails at such basic common sense? AI today is unbelievably intelligent and then shockingly stupid. It is unavoidable side effect of teaching AI through brute force scale.

Some scale optimists might say, don’t worry about this, all of this can be easily fixed by adding similar examples as yet more training data for AI. But the real question is this. Why should we even do that? You are able to get the correct answers right away without having to train yourself with similar examples.

Exercise №4 (⏳7 min to complete. 24 out of 45)

Watch the second part of the talk (⏳02:57 – 05:48 min), complete the sentences with the following words: aced, presumably, suspended, nonsense, measure, jug, randomly

A different one. I have a 12-liter jug and a 6-liter , and I want to ____ 6 liters. How do I do it? Just use the 6-liter jug, right? GPT-4 spits out some very elaborate __. Step one, fill the 6-liter jug. Step two, pour the water from 6- to 12-liter jug. Step three, fill the 6-liter jug again. Step four, very carefully pour the water from 6- to 12-liter jug. And finally, you have 6 liters of water in the 6-liter jug that should be empty by now.

OK, one more. Would I get a flat tire by bicycling over a bridge that is __ over nails, screws and broken glass? Yes, highly likely, GPT-4 says. __ because it cannot correctly reason that if a bridge is suspended over the broken nails and broken glass, then the surface of the bridge doesn’t touch these sharp objects directly.

OK, so how would you feel about an AI lawyer that _ the bar exam yet ___ fails at such basic common sense? AI today is unbelievably intelligent and then shockingly stupid. It is unavoidable side effect of teaching AI through brute force scale.

Не упускаем из внимания важный культурный аспект, которому хотим научить, а именно, вынудить задуматься над тем, что такое здравый смысл. Для этого мы прекращаем слушать, даже при условии, что у нас еще осталось половина видео. Оставляем вторую половину на следующий раз. И переходим к части, где задаемся вопросами и ищем самостоятельно ответы. Записываем мысли и готовимся представить наше мнение. Бесспорно, критическому мышлению, по мнению профессора Виллингама³ научить невозможно, но научить думать на каком-то материале, вполне доступно каждому учителю.

Мы будем размышлять, что такое здравый смысл. Для этого сначала ответим самому себе, что это такое и почему машина допускает глупые ошибки, а потом прочтем один или несколько комментариев из под видео и “подумаем” об них.

Critical thinking. Common sense puzzle

Exercise №5 (⏳4 min to complete. 28 out of 45)

Judging by the answers given by an extremely intelligent machine to a shockingly simple questions — questions even a primary school pupil could manage — let’s work in pairs or small groups to discuss what it lacks. How do you define common sense? Do you possess it? Why or why not?

Следующим этапом мы попросим учеников выразить мнение на короткий комментарий, оставленный под этим видео:
Exercise №6 (⏳4 min to complete. 32 out of 45)
Read the comment about the TED talk. Work in pairs or small groups. Do you agree with the viewer’s comment? Do you have anything to add?

I’m in love. This is so aligned with what I have learned to see from our past and current situation. The future situations are also clear. We are playing with potential dynamite using oven gloves and plastic googles, hoping that all goes well. The economic, and geopolitical motives for that are clear, but to what end. In the meantime, we need to ensure that we don’t let the genie out of the bottle and I believe the team developing AI and the concerned governments are doing the right things to protect everyone. We all need to be aware however, so that we can all participate and help to mitigating the situation, and advancing at the same pace as we are. A fine balance, but one that can be done as we have more than enough smart people in the world.

По окончанию каждого занятия у ребенка должен остаться артефакт его деятельности. В двух, прослушанных фрагментах, мы поговорили о то что такое здравый смысл, почему машине его не хватает. В комментарии об который им придется подумать затронут вопрос этической стороны вопроса. Остается попросить их презентовать собственное мнение о том что такое здравый смысл и почему машина не обладает им. Однако просто сказать “представьте личное мнение” недостаточно. Нужна инструкция. И снова на помощь приходит ИИ, мы попросим его написать небольшую инструкцию о том как составлять простую, пятиминутную презентацию.

Запрос: I need a short instruction for an English-speaking classroom on how to prepare and deliver a simple presentation on the topic of common sense. What is it, why intelligent machines don’t possess it, and I need a step-by-step instruction.
Ответ ChatGPT: In this English-speaking classroom, your task is to prepare and deliver a simple presentation on the topic of common sense and why intelligent machines lack it. Start by defining common sense and its key characteristics, then explain the reasons machines lack it, such as the absence of emotion, limited contextual understanding, and the inability to adapt. Illustrate the differences through examples and present contrasting cases where machines excel. Conclude by emphasizing common sense as a uniquely human quality and encourage thoughtful reflection. Keep the presentation engaging, concise, and use visuals to enhance understanding.

Зеленый текст можно использовать как простую инструкцию, и помощь в подготовке презентации.

Presentation skills

Exercise №6 (⏳12 min to complete. 44 out of 45)

You are going to give a short introduction to a talk “Why intelligent machines lack common sense”. Work in pairs or small groups. Use ideas bellow or your own ideas. Make a brief notes for an introductory paragraph and practice presenting your introduction aloud.

• Present real-life examples of AI misunderstandings.
• Explore challenges in AI’s understanding of simple things.
• Compare human and machine decision-making processes.

В этом видео еще осталось около 6 минут не отработанного материала, его можно или на персональную отработку или в качестве материала для следующего урока. Однако во второй части поднимаются достаточно сложные вопросы, которые имеют отношение к здравому смыслу, детей необходимо подготовить к поиску ответов на вопросы. Поэтому мы можем предложить самостоятельно посмотреть еще одно видео, которое также было найдено ИИ

, а на выходные посмотреть крайне интересный, документальный филь о создании AlphaGo.

References

2. 1. Евдокимова М., Агамалиев Р. Лингводидактический потенциал систем искусственного интеллекта // Экономические и социально-гуманитарные исследования. 2023. № 2 (38). C. 173 —191.
3. 1. Willingham D. Critical Thinking Why Is It So Hard to Teach? // Arts Education Policy Review. 2010. (109).