跟讀練習: 歐巴馬談 AI：喊放慢的公司「是真的在擔心自己的產品」【AI 大人說】 - 通過影片學習英語口語

選擇練習模式

每種模式都會開啟全屏練習室。

154 句

如果句子過短或過長，請點選 Edit 進行調整。

This conversation is in the United States of the United States of America.

It's the year 9th of the family, Joe Moore.

The table is Obama.

It's only AI this topic.

He talked about 20 minutes.

He said that the last call to be able to slow down the AI companies are really worried about their products.

But to be able to control AI, it's ultimately to be靠 the government.

The topic is the director of the director.

AI this tool every day changes.

It's in the number of the私人 companies.

How do you control the society?

Okay, settle in here.

I might have to...

I'm a former senator, this might feel like a filibuster, but I'm gonna take some time on this and I wanna go a little deep on it.

I do not believe that this technology is overhyped.

The commercial benefits of it may be overhyped.

You know, the valuations of AI companies may be overhyped.

The technology itself is not overhyped.

I've been tracking this for a long time I actually in

2015 -16 in the last couple years my presidency I actually

was focused on this pulled together a commission of the best

minds in this area to talk about the not just the technology

but also the political social economic ethical concerns around this produced a terrific report.

Apparently, it got put in the same bin as my pandemic proposal.

It wasn't picked up by the administration that followed me.

So we could see this coming, but part of what makes this profound is it accelerates.

so what is happening right now is we have reached this point of what's called recursive learning where essentially

and we had anticipated this

but we're now seeing it the machines are starting to be

able to teach themselves they don't have to wait for humans to start figuring things out

And so if you think about it as, let's say, humans are teachers and the AI models are students.

As recently as, let's say,

a year ago, 90 % of the learning of these models was given to them by the teacher, humans

and maybe five to 10 % they could fill in some gaps and figure stuff out.

In less than a year maybe even less than six months we're now at a stage where it's 50 -50

and if the trend lines continue the expectation is it might be 90 -10.

What what this means is that we're now

if the learning track of AI has been like this we're

now hitting the hockey stick trajectory where the models are going to get smarter

and smarter and better and better at a much faster pace, at an exponential pace.

What are the risks of that?

There's the big science fiction risk of,

wow, these models get smarter than us and they decide humans are fine, but not necessary.

They start setting their own goals and the killer robots kill us or we bow down to them.

I do not want to exaggerate that particular risk, but I will say there is a non -zero risk of the non -zero chance of

that happening but that's not actually the risk that I'm most concerned about although it's the risk that gets most attention

and the reason that's a risk is is not

because the the computer models are conscious necessarily it doesn't mean

that they are necessarily feel malice towards humans it's just

that if they start setting their own agendas they you may get a misalignment between what they want to do

and what we want them to do and

that gap can be dangerous that's problem number one the more serious problem is these models are getting powerful enough that

if they get in the hands of bad humans they can

do bad things they can be weaponized in certain ways they can do a lot of mischief That's a much bigger danger,

and if I said the danger of the first, or what's called the singularity,

and, you know, super intelligence taking over, if I say that's still a relatively small chance of that,

there's a great likelihood that rogue AI can cause a lot of havoc.

It's moving so fast that what happened over the last couple of weeks is that that a couple,

the leaders of what are called the frontier companies, the folks who are developing the very cutting edge of these models,

got scared enough themselves that they said, you know what, we need to slow down.

Now, two of those companies,

Anthropic and OpenAI, are about to launch massive IPOs that make all their employees millions and billions of dollars,

it is unprecedented for companies in that position to say,

you know what, our stuff is a little bit dangerous and maybe we need to stop or slow it down, not stop, but slow it down.

I say that because I think there are some people who think and understandably think, well these tech companies they might just be trying to block

out competition by saying we need to put a slow this down etc the mistrust of these big tech companies

and the amount of power that they have amassed I understand I will say

that in this situation there they are genuinely worried about their

product all right so now that I've got your attention the good news is is

that these are still these AI is a tool it's a machine it's built to give you the illusion

that it's not but it is it's a really sophisticated remarkable technology

but it it should be thought of as a tool which means we have choices in terms of how we use this, and you're right.

If we use it properly, if we direct it at drug development to cure cancer,

it could vastly accelerate our ability to rid the world of cancer.

If we train it and say, figure out zero carbon emission energy sources,

it could come up with a safe formula for nuclear fusion

that actually gives us much more plentiful energy without warming the planet.

Really useful stuff.

But if we get this wrong, it could be chaotic and potentially catastrophic.

So, my view this week has been it is good

that some of the leading companies have said we need to slow this down,

and we should encourage that.

There have been some on the left who have said we shouldn't allow them to make that decision.

I agree that

that can't be a long -term solution government has to be

regulating this I heard one of Donald Trump's main advisors on this make the argument

that the market will take care of it these companies will solve the safety issues

because they have very incentive to do so

if it turns out to be dangerous people will just sue them and and you You know, they're worried about financial liability.

That's not how we treat airlines, or drug companies, or food companies.

We now have 150 years of experience

just letting the market do whatever it wants when it comes to things that deeply affect our health and our safety.

And so there's no replacement for an effective government regulatory structure.

And the challenge we have is that obviously we have a president currently that has said that's for losers.

So we've got a gap at least in a couple of years where things are moving really fast

and we have an administration that is not capable of or willing to,

I would argue both, of putting together a serious regulatory framework,

which is why I actually think we should be encouraging some sort of voluntary industry restraint, not because it's a replacement,

but because that's probably, that may be the best we can do for now until we can get Congress and the White House

100

to start getting serious about this.

101

I know some states are going to be trying to initiate regulatory stuff

102

but it's hard to do this state by state just like gun control you know in my home state of Illinois

103

you have some strict gun control laws but the surrounding states do not

104

which means that they don't work this is more dangerous and it moves fasterAI自己想做的事跟人要他做的事对不上AI圈把这个叫做对齐问题就是他前面排第一个的那种风险

105

There is another alignment problem,

106

though, and that is the misalignment between how this technology can be best used for society

107

and the commercial imperatives of these companies.

108

we are these companies have attracted more capital

109

and our it's a bigger spend than the moon pro the

110

Apollo program it is it is one of the biggest

111

if not the biggest investment in infrastructure and science and technology in human history and it's happening very fast and

112

investors want to return and what that means is is

113

that there is pressure on these companies even if they're worried

114

that they have to come up with models that will justify all

115

that investment so the big thing right now in AI is what They're called agents, agentic AI.

116

You guys, those of you who've watched like Iron Man, right?

117

Iron Man's in his little suit, and he, I don't remember what the guy's name is, Jeeves or whatever, and he just talks, and the guy's like, he has a British accent or something,

118

and he kind of goes, yes, sir, you know?

119

And he solves all of these problems just with voice orders.

120

Then no, or if you've seen her, Scarlett Johansson, a very cool, you know, agentic AI.

121

That's the direction we're going in,

122

because the notion is that agentic AI will allow these AI bots to operate,

123

these models to operate in the world.

124

Be your assistant, replace the law associate or the accountant.

125

the problem is when you

126

if that's your business model you want to then get it out

127

and sometimes you have to test it by letting it run wild in the world to see

128

if it's actually working and that's where a lot of the danger may come in

129

because you're not in a sealed box it's it's in the internet it's looking at your bank accounts, et cetera.

130

I use that as an example to say,

131

if we are thinking about AI just in terms of how do we cure cancer or get better energy,

132

you can do that without having agentic AI and having it just roaming free in the internet.

133

The reason you are doing that is because you have to market a product

134

that people will pay money for that's a misalignment between what our society needs and

135

the commercial imperatives that these companies are facing not because necessarily they're trying to

136

do bad things but because you know they've got a they've got to justify these valuations so

137

that's one more reason why it is really important for us to have a competent government

138

and a serious bipartisan conversation around this issue, and we have to do it fast.

139

And I would encourage voters to pay attention to this.

140

If somebody does not have a serious plan for how to deal with this, then they're not meeting the moment, and you should probably look for somebody else.

141

All right.

142

Sorry about that, but I wanted to...

143

All right.

144

It's not the only thing we got to deal with, but it's a big one.

145

Obama said AI is a tool, how to use it for people who choose.

146

So he finally put the topic back to the選民.

147

The候選人 did AI拿不出认真的計劃, just換一個人.

148

In the video, Obama said, Obama said if he wanted to help him, he said he said he was a group of people who don't want to do it.

149

He said he said he said he said he said he

150

said he said he said he said he said he said

151

he said he said he said he said he said he

152

said he said he said he said he said he said

153

he said he said he said he said he said he

154

said he said he said he said he said I think it will be more likely to be more likely.

下載應用

流利口語所需的一切

AI 發音評分逐句打分

音標練習掌握每個音

詞彙學習建立你的詞庫

詞彙遊戲邊玩邊學

掃碼

Download on theApp Store Get it onGoogle Play

FOR YOU

本課的詞彙與口語要點

這段影片共有 154 個句子、2016 個單詞可供跟讀。講話部分時長為 15:58。說話人語速平穩，每分鐘約 126 個詞，很適合跟讀。 88% 的單詞屬於英語最常用的 3,000 詞，其餘的詞建議在練習前先學一下。

影片中的重點詞彙

影片中 15 個值得學習的單詞，附發音和釋義：

單詞	發音	釋義
topic 名詞	/ˈtɑpɪk/	主題 /主題, 話題 /話題
necessarily 副詞	/ˌnɛs.əˈsɛɹ.ə.li/	必然地, 必定地
encourage 動詞	/ɪnˈkʌɹ.ɪd͡ʒ/	鼓勵 /鼓勵, 激勵 /激勵
solve 動詞	/sɒlv/	解決 /解決
pace 名詞	/peɪs/	步伐, 步幅
cure 動詞	/kjɔː/	治療 /治療
justify 動詞	/ˈd͡ʒʌstɪfaɪ/	辯解 /辯解, 辯護 /辯護
accelerate 動詞	/ɪkˈsɛl.əˌɹeɪt/	加速
imperative 形容詞	/ɪmˈpɛɹ.ə.tɪv/	迫切的, 必須的 /必須的
gap 名詞	/ɡæp/	間隙 /間隙
killer 名詞	/ˈkɪlɚ/	殺手 /殺手, 兇手 /兇手
proposal 名詞	/pɹəˈpoʊzəl/	提議 /提議, 建議 /建議
argue 動詞	/ˈɑɹ.ɡju/	辯論 /辯論, 爭辯 /爭辯
infrastructure 名詞	/ˈɪnfɹəˌstɹʌkt͡ʃə/	基礎設施 /基礎設施
formula 名詞	/ˈfɔː.mjʊ.lə/	公式

影片中出現的短語動詞

單詞	釋義
come up with 動詞	想出, 提出
slow down 動詞	減速 /減速, 放慢
figure out 動詞	弄清楚
pick up 動詞	拾取, 撿 /撿
put together 動詞	拼合, 組合 /組合
take over 動詞	接管, 接手

影片中的語法

說話人最常用的結構，並附上影片中的原話：

結構	影片中的用法
現在完成時 have/has + 過去分詞 — 過去發生但與現在仍有關聯的事	have reached · has been · have amassed
被動語態 be + 過去分詞 — 強調發生了什麼，而不是誰做的	is overhyped · may be overhyped · is not overhyped
定語從句 who / which + 從句 — 補充說明人或事物	people who think · tool which means · people who choose

需要注意的發音

說話人用了 32 次縮略和弱讀形式，例如 they're, we're, don't。請按聽到的簡短形式來說。

“th” 音: ethical /ˈɛθɪkəl/, anthropic /ænˈθɹɒp.ɪk/
“sh” 和 “zh” 音: valuation /ˌvæ.ljuːˈeɪ.ʃən/, fiction /ˈfɪkʃən/, essentially /ɪˈsɛnʃəli/, associate /əˈsoʊʃi.ət/, potentially /pəˈtɛnʃ(ə)li/
長單詞——注意重音位置: necessarily /ˌnɛs.əˈsɛɹ.ə.li/, regulatory /ˈɹɛɡ.jʊ.lə.t(ɔː.)ɹi/, valuation /ˌvæ.ljuːˈeɪ.ʃən/, accelerate /ɪkˈsɛl.əˌɹeɪt/, imperative /ɪmˈpɛɹ.ə.tɪv/

如何用這段影片練習

先完整聽一遍影片，不要開口，記下不認識的單詞。
用正常速度逐句跟讀，每句重複到你的節奏與說話人一致為止。
錄下自己的聲音並與原聲對比，特別注意 topic, necessarily, encourage 這類單詞。

什麼是跟讀法？

跟讀法 (Shadowing) 是一種有科學依據的語言學習技巧，最初開發用於專業口譯員的培訓，並由多語言者Alexander Arguelles博士普及。這個方法簡單而強大：您在聽英語母語原聲的同時立即大聲重複——就像是一個延遲1-2秒緊跟說話者的影子。與被動聽力或語法練習不同，跟讀法強迫您的大腦和口腔肌肉同時處理並模模擬實的講話模式。研究表明它能顯著提高發音準確性，語調，節奏，連讀，聽力理解和口語流利度——使其成為雅思口語備考和真實英語交流最有效的方法之一。

影子跟讀法: 閱讀完整分步指南 →

跟讀練習: 歐巴馬談 AI：喊放慢的公司「是真的在擔心自己的產品」【AI 大人說】 - 通過影片學習英語口語

選擇練習模式

下載應用

推薦影片

‘We don’t have months’: Bernie Sanders sounds alarm on AI's 'extinction' threat

十年後的電腦會長什麼樣？黃仁勳：就像今天的手機，最少拿來打電話｜中英字幕

AI Governance Explained: Ethical, Transparent & Accountable AI for the Future

Jack Dorsey’s New Dessa AI Voice Sounds Scary Human

Alexandr Wang on what matters in an age of abundant intelligence

Anthropic just proved AI isn't getting better

Significant slowdown in AI payoff could tip economy into a recession, says Apollo Global's Slok

AI & Inequality | Daron Acemoglu: Big Tech poses risks as AI reshapes society

Nvidia’s Jensen Huang on the AI revolution, job losses and what drives him | Full interview

What AI can (and can't) currently do in a real lab | Jennifer Doudna

A $6.3 billion open-weight model just got embarrassed by the French...

AI Will End Philosophy, Physicist Claims