シャドーイング練習: B2 Level Discussion! Words, Listening and Debate | The Problem With AI Slop - YouTubeで英語スピーキングを学ぶ
B2
Hey, everybody.
304 文
文が短すぎたり長すぎる場合は、Editをタップして調整してください。
1
Hey, everybody.
2
Welcome back to High Level Listening.
3
Today's lesson is for those of you who are ready to move beyond everyday small talk and start having more real,
4
modern, and slightly deeper conversations in English,
5
especially if you're in the B1 level and you really want to push through and achieve that B2.
6
Yeah, that's exactly what we're working on today.
7
This is a B2 level conversation,
8
which means you'll hear more nuanced opinions,
9
more specific vocabulary, and maybe some ideas that are a bit more complex or abstract.
10
But this will still be a natural conversation.
11
And we think this is a great way to pick up new words
12
and phrases about topical issues with subtitles and captions to help you,
13
plus the chance to pause,
14
rewind and listen again so you don't get lost.
15
We also have all of our B2 level conversation study guides
16
that will help you stretch from small talk to deeper topics
17
in our conversational English study guide in the link in the description below.
18
That has PDF transcripts, dictionaries,
19
and so you can really feel confident in today's lesson.
20
Yes, so today's topic is something that everyone is talking about right now.
21
It's AI, artificial intelligence.
22
And we'll also cover AI-generated content.
23
Yeah, it's funny, kind of over the past couple of years,
24
I think there's been a lot of confusion in kind of the generic name AI.
25
And there's definitely been a shift in the way that most people use the word in day-to-day language.
26
It's starting to be more of an umbrella term
27
or kind of a catch-all phrase for all sorts of words and technology that we've used in the past,
28
even if those things aren't actually AI.
29
Words like CGI and computer animation,
30
you know, kind of like the computer graphics.
31
Some styles of cartoons, drawings and styles are also being called AI,
32
even if a real person manually or physically created them.
33
Yes, I know what you mean.
34
It's taking over the phrase.
35
When we looked at a photo that had been changed or manipulated,
36
we used to say, oh,
37
it's Photoshopped or someone Photoshopped this.
38
But nowadays, I think people say, oh, it's AI.
39
Oh, that's just AI.
40
It's fake.
41
yeah and another example is chatbots
42
if you need help from your bank your phone company
43
and you need to talk to someone you probably won't chat to a real person you'll chat to AI
44
or to say oh they've replaced the customer service team with AI oh I talked to the AI
45
and it didn't help me so it's replacing everything I know it's kind of weird it's
46
so the the term AI just kind of has lots of different meanings now.
47
Even anything that's fake on the internet,
48
fake news, disinformation, misinformation, people are in general,
49
they're just calling it AI or my new favorite phrase from the past couple of years, AI slop.
50
And you might have seen this as a comment.
51
Is this real or is this more AI slop?
52
Yeah, slop is that low quality,
53
obviously AI slop videos, pictures.
54
So yeah, there's obviously a lot of negativity around AI.
55
And we'll dive deeper into that topic and talk about AI-generated content or the slop.
56
And yeah, we've structured today.
57
So we'll start with sharing real experiences.
58
And then we'll talk about if we agree or disagree with opinions we've heard with a little debate.
59
And then share some deeper insights with articles that we've read.
60
And by the end, hopefully you'll feel a bit more confident discussing this topic naturally using B2 level English.
61
Okay, so we're going to start by sharing some of our own opinions.
62
We were talking about this with a couple of friends the other day,
63
and one of them made a really interesting point that kind of stuck with me.
64
She said that it's getting harder to tell what's real anymore.
65
It's getting harder to tell what's real anymore.
66
and that really succinctly captured something I think a lot of people are starting to notice.
67
She explained a little bit more that when she scrolls online now,
68
everything feels slightly off, like it's been polished and optimized,
69
but not genuinely thought through or the details are a little off.
70
It's not obviously bad, like maybe the very first look,
71
but it doesn't really feel fully authentic either.
72
She kind of went on to say that she'll read a post
73
or watch a video that sounds and looks really convincing at first,
74
but then it kind of falls apart the more you think about it or the more you look at it.
75
The ideas are there again,
76
the big picture is there.
77
The structure is there, but there's no real depth behind it.
78
And it's something kind of creepy,
79
a little bit weird, maybe even awkward.
80
And nothing really makes it memorable or worth coming back to.
81
It's just there.
82
It just exists.
83
It feels generic, kind of repetitive and kind of hollow in a way, really empty.
84
Like it's been generated for the sole purpose to imitate rather than actually provide information or education.
85
I think she kind of described it as a constant stream of content that looks good on the surface,
86
but really lacks substance underneath.
87
Yes, there is a name,
88
I remember, for that feeling when you watch something that's robotic or fake or digital,
89
and that strange feeling you get,
90
it's called the uncanny valley.
91
Oh, that's what that means real okay yeah
92
but you know it's not something's wrong something's missing like
93
when they make it I don't know what it is I can't pinpoint it exactly okay uncanny uncanny valley yeah
94
that feeling but yeah uh if we're talking about pictures and
95
and images uh videos especially they're not perfect
96
but they're already convincing enough to trick a lot of people including my mom and the older generation.
97
My mum sends me videos with lots of laughing emojis,
98
and the videos she sends me are blatantly AI.
99
There's artifacts in the videos.
100
Artifacts are those like glitches where somebody's got six fingers on one hand,
101
and she sent me one of all these cats in someone's bed,
102
and there's a cat with two heads,
103
and there's a cat that's extremely large compared to the other ones,
104
I think you sent it to me and she's petting it the wrong way as well.
105
Yeah, right.
106
So there's all these giveaways or these little artifacts or glitches.
107
But to the untrained eye or just at a glance, it's real enough.
108
And when people are scrolling on their phones,
109
they just look for a few seconds anyway.
110
And yeah, it's only when you look closer that you start to see the weird glitches.
111
And that's fine when it's just an innocent, silly video.
112
but I think with older people it's only going to get worse.
113
It worries me about my mum because if that fools her and it's quite rudimentary now,
114
as AI gets more and more sophisticated,
115
more lifelike and more believable,
116
she will have no idea if what she's seeing is real or not.
117
And even I'm starting to struggle with some things as well.
118
Even with the news.
119
I remember there was some news in Mexico recently of an airport on fire.
120
That was fake.
121
It wasn't a real video,
122
but enough people believed it for it to circulate through news channels and caused panic.
123
And it's only going to get worse as AI evolves and gets even better.
124
So yeah, I like to think of myself as someone who can spot AI pretty easily.
125
But even for me, I'm starting to doubt myself and it's starting to get harder.
126
So yeah, I agree.
127
It's getting harder to tell what's real anymore.
128
So our second point that we kind of wanted to make was that people are overreacting.
129
AI is just a tool.
130
So it just really depends on how it's being used.
131
Yeah.
132
So again, we were talking about this with our friend and he had a slightly different perspective.
133
It's worth mentioning that this comes up quite a lot in these discussions.
134
He basically argued that people are overreacting and it's just a tool.
135
It is something to assist you and help you.
136
If people are using it to produce low quality content,
137
then it's not really the fault of the technology,
138
but it's the person using it.
139
And he said that it can also be quite helpful in certain contexts,
140
like you can use AI to summarize information,
141
to improve clarity if you're writing something,
142
or help you organize your thoughts if you're trying to plan something,
143
which I think is a fair point.
144
I do have a friend who works as a software developer and he uses AI as well,
145
but he embraces it as a tool.
146
It helps him code faster,
147
but he still has to supervise it because the AI still makes mistakes.
148
So he's kind of got the best of both worlds.
149
He's got the speed of the AI and he's got his own expertise.
150
And importantly, it hasn't made him obsolete.
151
He hasn't lost his job because AI has taken it from him.
152
His knowledge is still critical.
153
I kind of see that,
154
but at the same time,
155
that's really mostly talking about the ones that are language models, right?
156
Like ChatGPT and Claude, right?
157
These are language models.
158
They're helping you interpret data,
159
summarize things, and help you speak more succinctly.
160
Okay, but it does feel like it oversimplifies what's actually happening
161
because the issue isn't just out the occasional user using ChatGPT for fun
162
or to summarize an email or a small business owner creating a post
163
or a social media post or something that benefits their small business.
164
What's really changed is the sheer volume and the speed at which this content is now being produced.
165
Nobody's fact checking.
166
Nobody's looking back to make sure that things are correct.
167
And a lot of it isn't really created to help the small business owner or help somebody in their job.
168
It's just meant to capture your attention.
169
It's clickbait, right?
170
So like, for instance, especially like in our job,
171
if you search something like B2 Level English on the YouTube search,
172
and you get search results that are pretty much 90 to 95% AI generated content with faceless videos,
173
AI generated pictures, and then on top of that,
174
AI generated voiceovers with zero real people involved.
175
And is that what people really want?
176
I mean, maybe a couple of those videos were popular at the beginning,
177
but do some people even realize that these people aren't real at all?
178
A lot of them are realistic photos.
179
Even the voiceover sounds okay.
180
And I bet if you're a non-native speaker,
181
you would hear it and think,
182
oh, maybe that's a real person.
183
But they're not.
184
I probably wouldn't notice if I wasn't currently aware of how much these videos are taking over the internet.
185
yeah i agree i think
186
that is a great example of the slop as well slop isn't just low quality
187
but it's an incredible volume
188
and a huge amount of videos like there are channels posting 10 videos every day
189
because you can just generate slop like it's just coming out of a pipe
190
and yeah i agree i I don't like it
191
when AI content is given the same platform as real people doing real work that's completely authentic.
192
So yeah, I guess you can try and control what you consume.
193
But how much responsibility can I have when I swipe to another video?
194
I have no idea if it's going to be real or AI.
195
I can't choose what goes on my feed.
196
So I don't like that.
197
Yeah, a lot of big platforms of prioritizing AI equal to real creators.
198
Okay, so to dig a little bit deeper,
199
we're going to bring in some articles and things that we've read.
200
This finally leads us into what was probably the most interesting point that someone brought up,
201
which is that it isn't just about content quality, okay?
202
It's not just about these little pictures and personal usage, right?
203
It's about how it's starting to affect the way that we think.
204
One of our more tech-savvy friends mentioned that when you're constantly exposed to this fast,
205
simplified, and slightly superficial content,
206
your expectations start to shift.
207
You start to lose your tolerance for anything that requires deeper focus
208
or sustained attention because your brain becomes used to these quick,
209
easily digestible, easily creatable information, right?
210
It felt like a subtle change at first,
211
but I think it really has a huge impact over time.
212
Now, I can say that I've definitely noticed that in myself personally,
213
especially when I try to read something more detailed or complex.
214
There's kind of this impulse to skim it,
215
just kind of skip ahead or just move on entirely,
216
just swipe through entirely.
217
Or worse, just use ChatGPT to summarize the information.
218
Almost every email platform has its own AI that will summarize information for you.
219
Now, I recently read this article in Time Magazine where they were reviewing a study done by MIT.
220
And basically what happened is they took a group of people ages 18 to 40
221
and split them into three groups to write a very formal style SAT exam.
222
That's kind of our high school exam that goes into college.
223
One group used ChatGPT, one used Google search,
224
and one used nothing at all.
225
And MIT researchers found, and I quote,
226
that of the three groups,
227
ChatGPT users had the lowest brain engagement and,
228
get this, consistently underperformed at neural, linguistic, and behavioral levels.
229
They went on to say that over the course of several months,
230
ChatGPT users got lazier with each subsequent essay,
231
often resorting to copy and paste by the end of the study.
232
Ouch.
233
And in just a few months, right?
234
I mean, ChatGPT, we've been using that for a couple of years now.
235
Imagine what it's doing to students who are even younger than us.
236
I mean, I'm already getting a little lazy.
237
Imagine what it's doing to students and people with no work ethic at all.
238
Yeah, clearly people starting to use it as a crutch,
239
something they can rely on.
240
Why do all the thinking when the AI can think for me. For me.
241
Yeah, right.
242
And I guess our brains are always looking for the laziest,
243
most efficient route to do things.
244
We call it efficiency, don't we?
245
Right.
246
Yeah.
247
That's what I call it.
248
I call it efficiency, but maybe it's being lazy.
249
Right.
250
Yeah.
251
It is slightly concerning to think if you're at school and you need to be developing your critical thinking skills,
252
you need to engage with your own thoughts,
253
not be spoon fed things.
254
But yeah, I can criticize it all I like,
255
but it's already affecting me.
256
I'm becoming more impatient in terms of doing research.
257
Sometimes I'll search for something in Google and if I don't get the answer immediately from Gemini,
258
ChatGPT, I'm like, oh, now I have to sift through lots of different links on Google.
259
It feels like such a chore to go out and do my own research.
260
So my tolerance for this deeper focus and sustained focus has definitely been lowered.
261
I know I can just ask ChatGPT or Gemini,
262
get an answer instantly, even if it's not perfect.
263
And then what's worse is that I've noticed that even the prompts that I write,
264
the sort of questions I ask ChatGPT,
265
those are getting simpler and lazier as well.
266
I remember when I first tried ChatGPT,
267
I would really like hold its hand and say,
268
alter the image so that the people are more realistic with wrinkles
269
and textures on their clothes but after using it for
270
so long and realizing how capable it is I just say make it more real
271
and it just does it so it's even affecting the way I talk to the AI is getting lazier because
272
sometimes I don't even correct spelling mistakes and it still gets what I mean which is as impressive as it is terrifying.
273
I think the other thing that's affecting the way I think is also affecting my vocabulary.
274
There are definitely words that I keep seeing from AI
275
and now I think they're starting to enter the lexicon generally as well.
276
So the word game changer comes up all the time now
277
because AI seems to use it a lot and I think people read it and then they start using it as well.
278
So I start seeing and reading it in articles.
279
And it makes me doubt whether it's written by a real person or whether it's AI.
280
Delve is another word that I didn't hear that often.
281
Then AI came out.
282
And then suddenly I see delve and dive into topics.
283
I see it in articles and in posts everywhere.
284
So it's even affecting the way we communicate,
285
or at least the way I do.
286
It's crazy.
287
Is AI affecting us or are we affecting AI?
288
I guess that's the big question now at the end of the day.
289
So there you have it.
290
There are definitely a lot of different perspectives on this and
291
that it's exactly what makes it such a good topic for our B2 conversation at this level.
292
Yeah, let us know what you think in the comments.
293
You don't need to express yourself perfectly,
294
but just try sharing your own experiences with AI or tell us
295
if you feel the same way as we do or
296
if you agree with any of the opinions you had in this video.
297
Yeah, and if you're tired of all the AI slop,
298
thank you for listening and studying with us here today.
299
We are absolutely real people.
300
Look, we have all 10 fingers.
301
AI can't do this.
302
AI can't do this, no. So we'll see you very soon for another episode here on High Level Listening.
303
Bye-bye.
304
Thank you.
アプリをダウンロード
話したすべての文をAIが採点
TRENDING
人気動画
このレッスンについて
このレッスンでは、日常の軽い会話から一歩進んで、より深く、現代的なテーマについて英語で話す練習をします。具体的には、人工知能（AI）やAI生成コンテンツに関する議論を通して、B2レベルの会話に挑戦します。このビデオでは、ニュアンスのある意見や具体的な語彙、少し複雑なアイデアを聞くことができ、字幕やキャプションが用意されているので、内容を追いやすくなっています。自分のペースで一時停止したり、巻き戻したりしながら、理解を深めることができます。
重要な語彙とフレーズ
- 人工知能（AI） - Artificial Intelligenceの略称。最近よく話題に上がる技術。
- CGI - コンピュータグラフィックスのこと。
- チャットボット - 顧客サポートなどで用いられる自動応答プログラム。
- フェイクニュース - 誤った情報やデマ。
- AIスロップ - 乱雑なAI生成コンテンツを指す新しい表現。
- Photoshopされた - 写真が加工されたことを示す表現。
- 情報の誤解 - 誤った情報の解釈。
練習のヒント
このビデオでは、速いスピードで自然な会話が行われるため、shadow speakを使用することが特に重要です。速いトーンに慣れるために、繰り返し聞いて、声に出してみてください。特に、AIに関する専門用語や表現に注目し、それらを自分の言葉で言い換えたり、応用したりする練習をすると良いでしょう。また、YouTubeで英語学習を通じてビデオを見直し、復習することで、理解が深まります。そして、IELTSスピーキング対策として、発音や流暢さを意識して練習することもおすすめです。この過程で、英語の発音を良くするための自信をつけることができるでしょう。
シャドーイングとは？英語上達に効果的な理由
シャドーイング（Shadowing）は、もともとプロの通訳者養成プログラムで開発された言語学習法で、多言語習得者として知られるDr. Alexander Arguelles によって広く普及されました。方法はシンプルですが非常に効果的：ネイティブスピーカーの英語を聞きながら、1〜2秒の遅延で声に出してすぐに繰り返す——まるで「影（shadow）」のように話者を追いかけます。文法ドリルや受動的なリスニングと異なり、シャドーイングは脳と口の筋肉が同時にリアルタイムで英語を処理・再現することを強制します。研究により、発音精度、抑揚、リズム、連音、リスニング力、そして会話の流暢さが大幅に向上することが確認されています。IELTSスピーキング対策や自然な英語コミュニケーションを目指す方に特におすすめです。