쉐도잉 연습: Anthropic is Eating The World - 영상으로 영어 말하기 배우기

연습 모드 선택

각 모드는 전체 화면 연습실로 열립니다.

158 문장
문장이 너무 짧거나 길면 Edit를 눌러 조정하세요.
1
Claude is not only eating your math homework, folks.
2
It is literally eating everything.
3
This started in 2025 when they launched Claude for Financial Services.
4
Then they launched Claude for Excel and Claude for Life Sciences.
5
In 2026, they launched Claude Coworks Legal Plug -in Suite.
6
You know what happened?
7
Right after, Thomson Reuters fell 16%.
8
Immediately after this, they launched Claude Code Security.
9
A few days later, CrowdStrike, big company, down 11%.
10
They've done the same thing with design.
11
I don't know if you guys remember Claude Design.
12
Figma fell about 7 % the day that that was released.
13
And the funniest thing about all this to me is that in the Anthropic filing for their IPO, they said, and I quote,
14
Anthropic may from time to time compete with enterprise software and application companies.
15
That sounds like code for we're coming for your ass.
16
Claude for small business.
17
Claude science.
18
They just did Claude for financial advisors, Claude Sheets, Claude Whiteboards, Claude Data, Claude Meetings, Claude for Manufacturing.
19
Where's Claude for doing my dishes?
20
How about the Claude for cleaning my toilet?
21
You know, I say this sort of tongue -in -cheek, but ladies and gentlemen, rest assured, they're going to get there soon too.
22
And it's obviously not just Anthropic.
23
It's obviously not just Claude.
24
Large language model companies and labs have been doing this thing for the better part of the last couple of years.
25
This month, OpenAI also launched ChatGPT for sales.
26
If you look on their little plugins library, it sits right next to Apollo, Clay, Gong.
27
These are pre -existing SaaS companies with a fair amount of usership.
28
You know, every day I'm scrolling through Twitter and I'm seeing somebody wake up and say, Claude just killed my startup.
29
OpenAI just destroyed my product.
30
You know, I think a lot of the time these were probably just wrappers to begin with.
31
But dude, if you look deep enough, isn't everything kind of a wrapper?
32
If you've ever wondered what's really happening here, it's that, you know, these AI companies,
33
they make models and other companies build products on top of those models.
34
Because other companies have built products on top of the models, the AI companies get to learn what sorts of things their users are asking,
35
and then they soak up all the IP.
36
And miraculously, the AI company typically starts to sell those products itself after a given amount of time.
37
On most plans, free or paid, your conversations, like your private one -on -one chats with Claude,
38
are actually training their models unless you turn that off.
39
For example, their consumer terms legitimately say it may use your chats to develop other products and services,
40
including training our models, unless you opt out.
41
And all model companies, including OpenAI, Google, XSI, and so on and so forth, all have the exact same settings.
42
However, what people don't understand is that is just about training the models.
43
Because even if you opt out, the companies are still going to learn how you use their products.
44
And ultimately, if you're working for another company that they may want to get into the industry of,
45
you know, ultimately speaking, what your SOPs and best practices are.
46
The only difference is they're not actually training the models.
47
Once every once in a while, Anthropic publishes this report called the Economic Index.
48
Now, in this process, Claude will read over 1 million business API transcripts
49
and then sort them by task and then do a bunch of analysis on them.
50
The analysis is specifically to determine what tasks Claude and AI are actually capable of doing, basically outsourcing, offshoring, or automating entirely.
51
So what does Claude and Anthropic more generally do with all of that analysis, aside from, of course, publish it in their economic report?
52
I think I have an idea.
53
And a really funny thing here is
54
that Anthropic's own terms actually say you may not use Claude to build a competing product,
55
but its filing says that Anthropic may compete with software companies.
56
So the rule is sort of one -sided.
57
And it makes sense that they do this, to be clear.
58
I mean, if I were a big daddy LLM company, I'd probably do the same thing.
59
By reading all your transcripts, the AI companies can legitimately see demand, both for the number of times their connectors or plugins are used,
60
but also more broadly, just what people are asking, what people want.
61
So that is the sure as shit clearest thing or signal that any company, let alone a large language model company,
62
can capitalize on to make a lot of money.
63
But at the same time, these guys also have distribution.
64
They're used by, you know, hundreds of millions, if not billions of people now.
65
They already have all the users.
66
They don't actually have to do any additional marketing.
67
They also have the advantage of being able to use AI to create products really quickly,
68
to then serve to the users that are on their AI platform.
69
And then they get to pitch it not as, oh, here's a new startup we're starting.
70
Literally, oh, here's just a one -click plugin that might be able to help you with that just a little.
71
You know, I'm not trying to say this is good or bad, but I am definitely trying to make sure you guys know what's going on.
72
Virtually all of these sophisticated intellectual property that you or your company,
73
dare I say, is currently feeding a clod in other models will soon be used against you.
74
I mean, it is just sort of the economically optimal thing to do, right?
75
If these large language model companies have all the data and the means to quickly replace you,
76
but even more broadly, if you're just using these products in your day -to -day work, eventually they're probably going to want to replace you, right?
77
And this is why I think a lot of companies are now starting to run open models.
78
In case you guys didn't know, an open wait model are models that you can download and then run right on your own servers.
79
So you're hosting it entirely yourself, meaning you're not actually sending like any API requests or data over to Cloud.
80
And the reason why that is important to companies is
81
because of knowledge alpha uh knowledge off is basically like your sops your prompts your customer data
82
you know information about your edge cases basically the stuff
83
that like you know your business has dutifully collected presumably over the course of the last few years
84
that make it unique and powerful
85
so at &t for instance something like 25 of all their ai usage currently right
86
now goes to open models their stated goals are to get
87
that to 70 to 80 i think they is like a big infrastructural company one
88
that probably deals with a lot of software are particularly worried about enterprise data leaking because as we know, enterprise data is a goldmine.
89
And I quote, and the tools are just a way to mine the gold.
90
They've also done lots of surveys on this.
91
At least 40 % of companies surveyed said that they want open weights models in production.
92
60 % of them have already moved some work off of current systems onto open weights models specifically to avoid this.
93
Now, you know, there are a lot of reasons why companies are doing this.
94
A big one is cost.
95
In addition to paying like the uh reaper's toll of all of your freaking data being leaked obviously
96
it costs a lot of money to use really intelligent models at scale uh
97
and when you buy your own compute and stuff like that
98
and actually have it run internally inside of an enterprise you get to take advantage of this batching effect
99
and basically save a fair amount
100
but more broadly yeah this is this is going on right now
101
and there's a reason why so many people are reaching out to me
102
and other people in the industry basically saying yo dude do you know anything about setting up these custom open weights models,
103
I don't think that we're comfortable sharing all this info with Anthropic, Big Daddy, OpenAI, etc. So what do you actually do with this information?
104
Well, the first thing is just don't build a freaking wrapper product.
105
I mean, god damn, are those wrapper products screwed.
106
This comes for everybody.
107
If you're going to build a product, it has to be a lot deeper.
108
You know, they just shipped personal finance, decks, you know, meeting notes,
109
sales prep, presumably like another dozen other products that I'm not even keeping up to date with.
110
But I mean, if you're making an app that just like visualizes slides, you're probably going to have to go a little bit deeper than
111
that if you don't want to get replaced literally the next couple of months.
112
The second thing is to build products and services just where these companies don't really think it's financially profitable to go,
113
at least not for the next few years.
114
I mean, Anthropics says they're going to compete in select categories.
115
What that means is they're going to compete in places where there is easy pickings and money to be made.
116
Maybe eventually over the course of the next decade, they'll get everything.
117
But at least in the short term, you know, pick a really narrow niche.
118
I'd also recommend physical work, something that intersects with the physical world is ideal.
119
How about regulated work, you know, work where somebody basically needs to be held accountable, businesses that deal with relationships,
120
vendors, sourcing, these sorts of things aren't very disruptible.
121
Another thing is to protect your SOPs.
122
So if you guys are going to use, you know, clod models or open AI models in order to do work, it's interesting.
123
I would probably not actually feed in my SOPs despite how compelling and simple it may be.
124
You know, it's sort of like the devil whispering on your shoulders, just paste that thing into the clod prompt.
125
Everything will be okay.
126
But the reality is, I mean, even if, you know, you feed the SOP once,
127
it'll be stored in a transcript that they could use at any point in time to rebuild that process.
128
So if your SOP is like super crazy dynamite, if that is really the moat of your business, be careful with that.
129
And then finally, use something that allows you to hot swap models.
130
Like I would keep open weights models for all sensitive work,
131
you know, assuming you are in a big enterprise with those proprietary SOPs I talked about.
132
And then, you know, just hot swap it to Claude In -N -Out
133
when you're working on something that maybe is a little bit less sensitive.
134
This is really annoying to do, to be clear.
135
But there are a lot of harnesses out there
136
that now offer this sort of thing is like a default option
137
and you can very quickly
138
and easily just switch between what model you want to use with these open harnesses
139
which make it super easy to just be like okay slash model let's change this okay feed
140
that in when you're done you know maybe you can open a new chat transcript
141
or something like that and then just type slash model
142
and go back to clod or chat gpt the way
143
that i see it these ai market opportunities um
144
that you know are currently being filled by these llm companies uh
145
you know they're they're one of many with ai being introduced to our economy, the frequency of these opportunities is going way up.
146
And so there are these vacuums that need to be filled that are constantly created by these,
147
you know, 10 ,000 model and product drops a day.
148
Also the magnitude of like total value you can exploit from one of these opportunities is also through the roof.
149
But the total amount of time that those opportunities are available, like in the case of a wrapper product, has also shrunk a lot.
150
And so, you know, AI companies are going to take the big obvious categories because it just makes sense.
151
But those like niche fast opportunities,
152
those ones that appear for like a month and then disappear that are typically tied to specific features or products dropped,
153
those ones are really ripe for the picking.
154
And I think anybody with like a growth hacker mindset can make,
155
you know, tens if not hundreds of millions of dollars over the course of the next year or two, if they take advantage of that.
156
So if you guys want help to find niches like that, just check out Makerschool.
157
That's more or less what we do.
158
Otherwise, please subscribe and I'll catch all y 'all in the next Sven.

앱 다운로드

유창하게 말하는 데 필요한 모든 것

AI 발음 평가문장마다 점수 확인
IPA 발음 연습모든 소리 마스터
단어 학습나만의 단어장 만들기
단어 게임놀면서 외우기

같은 채널

FOR YOU

추천 영상

이 레슨의 어휘와 말하기 포인트

이 말하기 레슨은 영상 “Anthropic is Eating The World”을(를) 바탕으로 합니다. 가장 자주 반복되는 단어는 다음과 같습니다: model, Claude, companies, product. 이 영상에는 섀도잉할 문장 158개와 단어 2157개가 있습니다. 말하는 구간의 길이는 9:54입니다. 화자는 분당 약 218단어로 빠르게 말하므로 연음과 약하게 발음되는 소리가 많습니다. 단어의 84%가 영어에서 가장 많이 쓰이는 3,000단어에 속합니다. 나머지는 연습 전에 미리 확인해 두세요.

이 영상의 핵심 어휘

영상에 나오는 익혀 둘 만한 단어 15개를 발음, 뜻과 함께 정리했습니다.

단어발음뜻
enterprise 명사/ˈɛntəˌpɹaɪz/사업, 기업
sensitive 형용사/ˈsɛn.sɪt.ɪv/민감하다
daddy 명사/ˈdædi/아버지, 아빠
publish 동사/ˈpʌblɪʃ/출판하다
leak 명사/liːk/누설, 루설
swap 동사/ˈswɑp/교환하다
prompt 동사/pɹɑmpt/부추기다
server 명사/ˈsɝvɚ/서버, 봉사기
select 동사/sɪˈlɛkt/고르다
ideal 형용사/aɪˈdi(ə)l/이상적, 리상적
custom 명사/ˈkʌstəm/습관, 풍속
pitch 명사/pɪt͡ʃ/투구
deck 명사/ˈdɛk/갑판
narrow 형용사/ˈnæɹəʊ/좁은, 좁다
collect 동사/ˈkɑlɪkt/수집하다

영상에 나오는 구동사

단어발음뜻
wake up 동사/ˌweɪk ˈʌp/눈뜨다, 일어나다

이 영상의 문법

화자가 가장 많이 쓰는 문형을 영상 속 실제 표현과 함께 정리했습니다.

문형영상 속 표현
수동태 be + 과거분사 — 누가 하는지보다 무슨 일이 일어나는지에 초점was released · are used · be used
현재완료 have/has + 과거분사 — 과거의 일이 지금도 관련이 있을 때They've done · you've ever wondered · have built

주의할 발음

화자는 you're, they're, don't 같은 축약형과 약화된 형태를 42번 사용합니다. 들리는 대로 짧게 발음하세요.

  • “sh”와 “zh” 소리: publish /ˈpʌblɪʃ/, dish /dɪʃ/, assure /əˈʃʊɹ/, offshore /ɒfˈʃɔː(ɹ)/, visualize /ˈvɪʒuəˌlaɪz/
  • 긴 단어 — 강세 위치에 주의: presumably /pɹɪˈzuːməbli/, intellectual /ˌɪntəˈlɛkt͡ʃʊəl/, intelligent /ɪnˈtɛlɪd͡ʒənt/, virtually /ˈvɚ.t͡ʃə.wə.li/, economically /ˌiː.kəˈnɒm.ɪ.kli/

한국어 화자가 어려워하는 소리:

  • /f/ — ㅍ(/p/)으로 바꾸지 말고 윗니를 아랫입술에 대기: frequency /ˈfɹiːkwənsi/, default /dɪˈfɔlt/, alpha /ˈæl.fə/, folk /foʊk/, profitable /ˈpɹɑ.fɪ.tə.bl̩/
  • /v/ — ㅂ(/b/)과 구별하기: sensitive /ˈsɛn.sɪt.ɪv/, server /ˈsɝvɚ/, devil /ˈdɛvəl/, virtually /ˈvɚ.t͡ʃə.wə.li/, vacuum /ˈvæ.kjuːm/
  • /z/ — ㅈ이 아니라 성대를 울리는 /s/: enterprise /ˈɛntəˌpɹaɪz/, presumably /pɹɪˈzuːməbli/, advisor /ædˈvaɪ.zə(ɹ)/, capitalize /ˈkæpətəˌlaɪz/, visualize /ˈvɪʒuəˌlaɪz/

이 영상으로 연습하는 방법

  1. 먼저 말하지 않고 영상을 끝까지 듣고 모르는 단어를 적어 둡니다.
  2. 0.75배속으로 한 문장씩 섀도잉을 시작하고, 익숙해지면 보통 속도로 돌아갑니다.
  3. 자신의 목소리를 녹음해 원본과 비교하고, enterprise, sensitive, daddy 같은 단어에 특히 주의합니다.

쉐도잉이란? 영어 실력을 빠르게 키우는 과학적 방법

쉐도잉(Shadowing)은 원래 전문 통역사 훈련을 위해 개발된 언어 학습 기법으로, 다언어 학자인 Dr. Alexander Arguelles에 의해 대중화된 방법입니다. 핵심 원리는 간단하지만 매우 강력합니다: 원어민의 영어를 들으면서 1~2초의 짧은 지연으로 즉시 소리 내어 따라 말하는 것——마치 '그림자(shadow)'처럼 화자를 따라가는 것입니다. 문법 공부나 수동적인 청취와 달리, 쉐도잉은 뇌와 입 근육이 동시에 실시간으로 영어를 처리하고 재현하도록 훈련합니다. 연구에 따르면 이 방법은 발음 정확도, 억양, 리듬, 연음, 청취력, 말하기 유창성을 크게 향상시킵니다. IELTS 스피킹 준비와 자연스러운 영어 소통을 원하는 분들에게 특히 효과적입니다.

섀도잉 방법: 단계별 전체 가이드 읽기 →