Практика Shadowing: ‘We don’t have months’: Bernie Sanders sounds alarm on AI's 'extinction' threat - Изучайте разговорный английский по видео
Загрузка...
1
A bombshell sudden departure from a top AI researcher is ringing alarm bells about the dangers of the industry, and they are being heard everywhere across the world and in Washington, D.C.
2
Today, Jacob Coxon announced out of nowhere that he is quitting his role training new AI models for Anthropic, citing his concerns that the lab and its competitors like OpenAI,
3
Google, and others are racing to build systems that they will not be able to control.
4
I was working on building AI systems to make them smarter.
5
And recently I've come to the realization that many people expect the AI systems in the next few years
6
to be potentially capable of essentially taking over the world.
7
This technology is not ordinary technology.
8
It's not like other forms of innovation.
9
This is essentially a super weapon.
10
We know how to control nuclear weapons pretty much.
11
We don't yet know how to control AI.
12
This is possibly the most dangerous technology that humanity has ever created.
13
People are begging for regulation.
14
I think it's just a matter of international somebody stepping in
15
and being able to ensure that we don't race ourselves into extinction.
16
That kind of dire warning has begun to capture the attention of politicians, many of whom are belatedly waking up to the dangers of this technology for the role,
17
the needed role of the government in some regulatory structure.
18
One of the politicians actually been out front on legislating this from early on is Senator Bernie Sanders.
19
And Senator Sanders, independent of Vermont, joins me now.
20
Senator, I know this is something that you've been spending a lot of time thinking about and talking about recently.
21
It seems like with this, there's a few things that have happened in the last few weeks.
22
This incident in which an open AI model was able to sort of break out of open AI
23
and get out of the Internet and commit a felony by hacking another company, which if a human did it, like that's a crime.
24
And then a bunch of these departures, including coxswains saying, this is going too fast.
25
We're going somewhere very dangerous.
26
My question for you is, OK, so what can be done?
27
What what should we do if you've got all these people saying this is not good?
28
Well, there's an obvious answer to that, Chris.
29
If all of these guys acknowledge they're building a dangerous technology and they don't know where this technology is going,
30
you don't have to be genius is to say hey stop
31
that development certainly we don't want to go into super intelligent
32
ai we got to ban that and we got to stop right now
33
and figure out how we make sure that ai
34
and robotics work for ordinary people and not just make mr musk
35
and other billionaires even reach up that ain't complicated
36
and in fact on thursday i'm bringing in some of the uh most uh knowledgeable people on this issue into Capitol Hill.
37
We've got a closed door meeting.
38
I invited the entire United States Senate, some House members.
39
Hopefully out of that can come some serious legislation to control AI, which is in fact what the American people want.
40
The stop, I mean, stop seems, you know, that seems simple enough.
41
Here's I want to read you a quote from from someone who replied to his ex-colleague Jacob Cox,
42
who I just played, who said he earnestly believes that people inside the labs earnestly believe that they might, you know, imperil all of humanity.
43
And this is someone, Anthropik, who says he's correct.
44
We really do earnestly believe AI could kill all humans.
45
I personally think it's greater than 10 percent within the next decade.
46
I believe Anthropik is trying its best, but we not yet have a plan to solve alignment for super intelligence and are not clearly on track.
47
So two things.
48
First is.
49
Why would you want to win a race off a cliff?
50
Right.
51
So tell them to stop.
52
Right.
53
Moratorium.
54
But again, that legislation takes a long time.
55
I guess my question here is, it sort of seems like we're at a tipping point now-ish.
56
Like, are you worried about the timescale that the political system works on and that the technology is developing?
57
Well, I'm worried about a lot of things.
58
I'm worried about the power of the AI industry, which is owned by some of the wealthiest people on Earth.
59
If you talked a moment ago about the role of money in politics, and we've got a corrupt campaign finance system,
60
the AI industry is prepared to spend hundreds and hundreds of millions of dollars to oppose any type of sensible legislation.
61
That concerns me.
62
And the other concern is that the political process moves very slowly under the best of circumstances.
63
So what we have got to do now is light a match under the backsides of members of Congress and say, you know what?
64
We don't have months.
65
We don't have years.
66
We got to move now.
67
And maybe the very first thing we can do is to convince our president, who appears to know absolutely nothing about AI and the dangers of AI,
68
to sit down with President Xi of China
69
and work on a moratorium on preventing the development of super intelligent AI
70
and a pause so that we don't get into real trouble for the future of humanity.
71
I'm glad you brought that up because it speaks to two things.
72
One is a lot of people say, well, if we don't do it, the Chinese or someone else will build it.
73
Right.
74
Which, again, OK, I mean, you know, if there's if there's two countries are going to create a weapon that extinguishes humanity, I think it's better if we go down to one.
75
Personally, it seems like we've gotten half we've sort of solved half the problem.
76
But but there is actually like an upcoming bilateral meeting between the president of China and the president of the US.
77
We're like, this will be a top agenda item.
78
And there could actually be some movement on this.
79
Again, this isn't like some abstract thing.
80
It's gonna happen the next few weeks.
81
Absolutely.
82
And look, think back in the 1980s where you had Gorbachev,
83
head of the soviet union ronald reagan a conservative president of the united states disagreed on everything
84
but they did understand that a nuclear war would not be good for either country
85
or for the world they reached a nuclear arms agreement and
86
that is exactly what we have got to do right now
87
and i hope that there are people in the trump administration who are prepared to push the president to do that.
88
And I will tell you this, that the polling is now off of the charts.
89
It's not only that people don't want data centers in their own communities.
90
They are very frightened about the inaction of Congress, and they want Congress to take action.
91
They're worried about the existential threat that these scientists and researchers are talking about.
92
They want us to act, and it's incumbent upon us to do just that.
93
I want to just leave you with the polling on that, just to fact check you here.
94
New polling out from research for progress, data for research, progress for progress research, sorry.
95
Voters who support regulating AI by party, Democrats, 72%, independents, 70%, Republicans, 63%.
96
So there really is a growing consensus on this.
97
That's not hyperbole, Senator Bernie Sanders.
98
Thanks for making time with us tonight.
99
I really appreciate it.
100
Good.
101
Thank you.
102
Hey, it's me, Rachel Maddow.
103
If you like the story you just watched, you should know about a new way to see even more
104
from my show and from all of MSNOW live or on tape.
105
It's the new MSNOW membership.
106
You can watch everything we do on your phone or your tablet or your computer.
107
Plus you can join conversations with me and the other hosts
108
and with other folks who care about the news as much as you do.
109
MSNOW membership.
110
During September, you can sign up for just 40 bucks a year, which is like just over three bucks a month.
111
Just click on the MSNOW logo right here or go to ms.now slash membership.
112
Hope to see you there.
О данном уроке
В этом уроке ты попрактикуешься в воспроизведении речи на английском, используя диалог из видео о проблемах ИИ. Ты улучшишь произношение английского, освоишь ключевые фразы и научился повторять речь за носителями с естественной скоростью. Это отличная возможность проверить, как ты справляешься с эмоциональной интонацией и быстрым говорением — именно такими навыками нуждаются все, кто учит английский по видео!
Ключевые слова и фразы
- Alarm bells — тревожные звонки (в переводе: "звонки тревоги", используется для обозначения опасности).
- Super weapon — сверхоружие (выражает идею чрезвычайно опасного оружия).
- Regulation — регулирование (ключевое слово в обсуждении контроля за технологиями).
- Alignment — согласование (в контексте ИИ: соответствие целей ИИ целям человека).
- Race off a cliff — бег к пропасти (идиома, обозначающая движение к катастрофе).
Советы для практики
Этот диалог отличается эмоциональной интонацией и быстрой речью, поэтому идеально подходит для shadowspeak (техники повторения за говорящим). Вот как практиковаться:
- Сначала прослушай отрывок по 5-10 секунд, затем остановь и повтори сразу. Старайся копировать не только слова, но и тон, ударения и паузы — это поможет улучшить произношение английского быстрее, чем ты думаешь!
- Используй slow-mo режим (если это возможно) для сложных фраз, например: "We don't have months" или "Why would you want to win a race off a cliff?". Повтори их до тех пор, пока не почувствуешь, что речь лёгка.
- Запиши себя на телефоне и сравни с оригиналом. Это поможет увидеть, где нужно подкорректировать интонацию или скорость. Наш сайт для shadowing — отличный инструмент для такой работы, но даже без него можно добиться прогресса!
Помни: практика shadow speak даёт быстрые результаты, если заниматься регулярно. Даже по 10 минут в день — и ты заметишь, как твой английский становится более уверенным и естественным!
Что такое техника Shadowing?
Shadowing — это научно обоснованная техника изучения языка, изначально разработанная для подготовки профессиональных переводчиков и популяризированная полиглотом доктором Александром Аргуэльесом. Метод прост, но эффективен: вы слушаете аудио на английском от носителей языка и немедленно повторяете вслух — как тень, следующая за говорящим с задержкой в 1–2 секунды. В отличие от пассивного прослушивания или грамматических упражнений, Shadowing заставляет мозг и мышцы рта одновременно обрабатывать и воспроизводить реальные речевые паттерны. Исследования показывают, что это значительно улучшает точность произношения, интонацию, ритм, связную речь, понимание на слух и беглость речи — что делает его одним из самых эффективных методов для подготовки к IELTS Speaking и реального общения на английском.