跟读练习: Will AI Take Your Job in the Next 10 Years? Wrong Question | Vinciane Beauchene | TED - 通过YouTube学习英语口语
困难
跟读控制
0% 已完成 (0/135 句)
Back in the 50s, Alan Turing came up with an idea.
⏸ 已暂停
速度:
重复次数:
等待模式:
字幕同步:0ms
所有句子
135 句
1
Back in the 50s, Alan Turing came up with an idea.
0:04.26 – 0:08.38 (4.1s)
2
If you couldn't tell if you were talking to a machine or a human, it meant the machine must be intelligent.
0:09.26 – 0:17.54 (8.3s)
3
He coined the Turing test.
0:17.58 – 0:19.34 (1.8s)
4
Today, most chatbots pass the test easily.
0:20.38 – 0:24.14 (3.8s)
5
But here's the catch.
0:25.14 – 0:26.98 (1.8s)
6
I believe the test was wrong.
0:27.02 – 0:29.10 (2.1s)
7
Because talking isn't what's going to change the world.
0:30.26 – 0:34.50 (4.2s)
8
Doing is. That's why I ask a slightly different question to the leaders I work with.
0:34.54 – 0:41.90 (7.4s)
9
On a daily basis, my role is to reshape organizations, trying to find the right mix of strategy, tech and talent.
0:41.90 – 0:51.66 (9.8s)
10
And my obsession is to make sure that talents do not get out of the equation.
0:52.30 – 0:57.78 (5.5s)
11
So the question I ask my clients is: if an AI could take over all of your team's tasks, who would you keep and why?
0:58.86 – 1:11.30 (12.4s)
12
That question is strategic.
1:11.34 – 1:13.78 (2.4s)
13
And the answer matters to me, not just intellectually, but because I have two daughters at home.
1:13.78 – 1:20.54 (6.8s)
14
They are five and nine.
1:20.58 – 1:22.34 (1.8s)
15
And right now, as you can see, they feel invincible.
1:22.38 – 1:26.66 (4.3s)
16
But I keep wondering: What is the world of work they will step into?
1:26.66 – 1:32.42 (5.8s)
17
We need to build a future where humans matter more, not less.
1:32.46 – 1:37.38 (4.9s)
18
Now let me try to illustrate how this is playing out in the field.
1:38.30 – 1:42.50 (4.2s)
19
A consumer goods client of mine is all in on AI.
1:43.50 – 1:47.74 (4.2s)
20
They didn't want to just deploy the next algo.
1:48.62 – 1:51.86 (3.2s)
21
They wanted to rethink the selling process itself.
1:51.86 – 1:56.22 (4.4s)
22
The trigger was agents.
1:56.26 – 1:58.66 (2.4s)
23
Have you heard about agents?
1:59.78 – 2:01.86 (2.1s)
24
They are the latest generation of AI: more autonomous, able to connect across systems, to plan, to take action, to learn, to adapt.
2:01.86 – 2:12.78 (10.9s)
25
The James Bond of AI.
2:12.78 – 2:14.58 (1.8s)
26
And applied to the selling process, you get an agent that is able to target the customer, make recommendations, negotiate, close the deal.
2:15.70 – 2:26.90 (11.2s)
27
All this with no human intervention.
2:26.90 – 2:29.50 (2.6s)
28
A fully autonomous sales engine.
2:30.38 – 2:33.42 (3.0s)
29
And it was technically feasible.
2:34.66 – 2:36.78 (2.1s)
30
But then an exec asked, "Hmm, if the machine does all of this, then what remains for humans?" This cracked everything open because when we looked deeper at their most loyal customers, we saw they weren’t sticking around because of prices or products but because of how the sales rep made them feel.
2:37.78 – 3:02.74 (25.0s)
31
So we flipped the model around.
3:03.74 – 3:06.34 (2.6s)
32
Humans were no longer going to be about pushing products.
3:06.38 – 3:10.26 (3.9s)
33
They were going to be about building relationship, belonging, loyalty.
3:10.30 – 3:15.90 (5.6s)
34
Very concretely, this meant new skills, new incentives, a very different mindset.
3:17.22 – 3:24.26 (7.0s)
35
Well it changed everything.
3:24.30 – 3:26.22 (1.9s)
36
But it worked.
3:26.26 – 3:28.90 (2.6s)
37
Because in the age of AI, human value isn't gone.
3:28.90 – 3:33.18 (4.3s)
38
It's just moved.
3:33.22 – 3:34.46 (1.2s)
39
Now I’m not talking about copilots anymore.
3:35.74 – 3:38.46 (2.7s)
40
For a while the narrative has been AI will augment us, not replace us.
3:38.50 – 3:45.06 (6.6s)
41
Well this is not where the tech is going today.
3:45.10 – 3:47.94 (2.8s)
42
And I believe we have real hard work to do if we want this narrative to stay true.
3:48.82 – 3:53.46 (4.6s)
43
So I'll say a few words about what I think needs to be done in a second.
3:55.50 – 3:59.74 (4.2s)
44
But first, let me tackle three myths that I think are holding us back.
3:59.74 – 4:04.58 (4.8s)
45
I call them "head in the sand" ideology.
4:05.82 – 4:09.02 (3.2s)
46
Number one.
4:10.26 – 4:12.02 (1.8s)
47
"All of this is overblown. We'll adapt." Yes. We've adapted to electricity, the industrial revolution, the internet.
4:12.06 – 4:23.82 (11.8s)
48
But we've done so on the back of generations that did not have the training nor the time to adapt.
4:23.82 – 4:29.58 (5.8s)
49
And in the case of this revolution, time is of the essence.
4:30.54 – 4:34.70 (4.2s)
50
You may think you have time because agents are just emerging.
4:35.66 – 4:39.10 (3.4s)
51
And it's a fact.
4:39.14 – 4:40.46 (1.3s)
52
Our research shows that today only 13 percent of companies have embedded agents in their workflows.
4:40.50 – 4:47.78 (7.3s)
53
But tech moves exponentially.
4:49.46 – 4:51.82 (2.4s)
54
Humans, they crawl linearly.
4:51.82 – 4:54.62 (2.8s)
55
If you don't prepare now, you'll struggle to keep up.
4:54.62 – 4:58.46 (3.8s)
56
And I'm not talking about science fiction.
4:59.26 – 5:01.94 (2.7s)
57
I'm not talking about AGI, artificial general intelligence, this moment where AI will be smarter than us.
5:01.94 – 5:09.98 (8.0s)
58
I'm referring here to ACI, artificial capable intelligence.
5:10.02 – 5:16.66 (6.6s)
59
The moment when AI will be able to take on ambiguous, complex goals with minimal oversight.
5:17.58 – 5:26.90 (9.3s)
60
And while AGI is speculative, ACI is a deadline.
5:26.90 – 5:32.62 (5.7s)
61
While we spend hours debating about superintelligence and consciousness, we miss the milestones that ACI is meeting with increasing frequency.
5:32.62 – 5:43.58 (11.0s)
62
ACI will change how work is done and by whom.
5:44.62 – 5:48.10 (3.5s)
63
Let's shape it, not wait and see.
5:49.06 – 5:51.50 (2.4s)
64
Now myth number two.
5:52.66 – 5:54.02 (1.4s)
65
"Soft skills are our sweet spot." Yes, it's lovely to believe that empathy, creativity are uniquely ours.
5:54.86 – 6:05.46 (10.6s)
66
But evidence says otherwise.
6:06.66 – 6:09.86 (3.2s)
67
More and more humans like to interact with AI because they feel it's more empathic.
6:09.86 – 6:16.90 (7.0s)
68
And why not?
6:18.46 – 6:19.74 (1.3s)
69
I mean, AI doesn't get tired, doesn't get cranky, doesn't judge you.
6:19.74 – 6:28.06 (8.3s)
70
So this moat we thought was ours, it's shrinking.
6:28.74 – 6:32.62 (3.9s)
71
And we need to stop asking what AI can't do and focus on where humans make a difference and why.
6:33.22 – 6:41.14 (7.9s)
72
At this stage, I'm sure you would love me to come up with the list of human qualities that will remain ours forever.
6:43.78 – 6:52.34 (8.6s)
73
But my point is, there is no universal list.
6:53.30 – 6:55.82 (2.5s)
74
Each company needs to figure it out based on its strategic positioning.
6:56.54 – 7:02.34 (5.8s)
75
This is hard, uncomfortable work, but it's work that you as leaders need to take on.
7:02.38 – 7:09.06 (6.7s)
76
Now myth number three.
7:10.34 – 7:11.74 (1.4s)
77
My preferred one. I'm French.
7:13.06 – 7:14.94 (1.9s)
78
"We need to protect jobs." Yes. I see where this one is coming from.
7:14.94 – 7:21.78 (6.8s)
79
Today, 41 percent of employees believe that their job will vanish in the next decade because of AI.
7:21.78 – 7:29.42 (7.6s)
80
But protecting jobs is like anchoring a boat in a storm.
7:30.62 – 7:34.78 (4.2s)
81
Jobs are fixed.
7:34.78 – 7:36.14 (1.4s)
82
The human potential to grow and adapt, on the other hand, is not.
7:37.06 – 7:40.94 (3.9s)
83
This is where we need to invest.
7:41.62 – 7:44.06 (2.4s)
84
The challenge is our organizations are not geared for that today.
7:44.10 – 7:48.02 (3.9s)
85
Org charts are static.
7:48.06 – 7:49.86 (1.8s)
86
Career paths are narrow.
7:49.86 – 7:51.78 (1.9s)
87
Training is occasional.
7:51.78 – 7:53.62 (1.8s)
88
This system will fall apart the day that the boundaries of jobs start melting away fast.
7:53.62 – 7:59.90 (6.3s)
89
So what do we need to do?
8:01.18 – 8:03.94 (2.8s)
90
Let me take you to an ideal company.
8:05.62 – 8:08.38 (2.8s)
91
Not a theoretical one, just the blend of the boldest clients I've worked with.
8:08.42 – 8:13.66 (5.2s)
92
First, they don't start with tech.
8:14.82 – 8:17.42 (2.6s)
93
They start with strategy.
8:18.62 – 8:20.02 (1.4s)
94
They focus on the outcomes that truly differentiate them on the market.
8:21.14 – 8:25.70 (4.6s)
95
They understand how agents will allow them to deliver against those outcomes in totally different ways.
8:26.38 – 8:33.50 (7.1s)
96
And they look at where people still make a difference for the better.
8:34.42 – 8:38.66 (4.2s)
97
As you can see, this is not incremental redesign of your operating model.
8:39.66 – 8:44.98 (5.3s)
98
It's radical AI-first reinvention.
8:45.02 – 8:48.34 (3.3s)
99
And we did this work for an industrial goods client of mine.
8:49.30 – 8:53.18 (3.9s)
100
Imagine having to go through 50 “hack a future” workshops, looking at how AI is going to disrupt each of your businesses, each of your function.
8:53.22 – 9:03.50 (10.3s)
101
Comfortable? It is not.
9:03.54 – 9:05.34 (1.8s)
102
But it allowed the leaders to align on a vision of where agents win, people matter and how best to pair them.
9:06.26 – 9:14.34 (8.1s)
103
Now once you have this vision, you want to translate it into a workforce model.
9:15.78 – 9:21.30 (5.5s)
104
How many people do I need?
9:21.34 – 9:23.26 (1.9s)
105
With what skills?
9:23.30 – 9:25.34 (2.0s)
106
No more guesswork, just informed, intentional reinvention.
9:25.38 – 9:30.34 (5.0s)
107
A multiyear skills forecast.
9:30.38 – 9:33.10 (2.7s)
108
And this is something we built for a consumer goods client that was facing a massive challenge.
9:33.14 – 9:39.42 (6.3s)
109
Imagine having to reformulate your entire product portfolio while keeping the leadership and innovation.
9:39.46 – 9:47.54 (8.1s)
110
Of course, AI unlocked the productivity that was required, but the work was much deeper.
9:47.58 – 9:54.10 (6.5s)
111
They needed to reinvent the role of the researcher.
9:54.14 – 9:57.14 (3.0s)
112
From chemist to data-driven biologist, from solo expert to multifunctional teammates.
9:58.06 – 10:05.74 (7.7s)
113
And they made it happen because they mapped very precisely the future skills that they needed, and they built a very effective upskilling and mobility engine.
10:05.74 – 10:15.50 (9.8s)
114
Then you want to publicly commit to taking your talents to their fullest potential.
10:17.10 – 10:24.02 (6.9s)
115
Now I know what you're going to tell me.
10:25.06 – 10:27.02 (2.0s)
116
Vinciane, why would we invest in talent if an AI can do their job faster, cheaper and without complaining?
10:27.06 – 10:37.38 (10.3s)
117
Well because the day that interacting with an AI becomes the new norm, a commodity, the interaction with humans is going to take an entire new meaning.
10:38.78 – 10:50.02 (11.2s)
118
Trust, authenticity, accountability.
10:50.62 – 10:54.42 (3.8s)
119
Those are the values we will anchor on.
10:55.46 – 10:57.86 (2.4s)
120
So the smartest companies will invest in talent.
10:58.94 – 11:03.22 (4.3s)
121
Not only tech talent, all talent.
11:03.26 – 11:06.14 (2.9s)
122
Not once, but systematically.
11:06.18 – 11:09.34 (3.2s)
123
And they will protect time to learn.
11:09.38 – 11:13.14 (3.8s)
124
Because today, while freelancers spend on average four hours per week learning, employees spend none.
11:13.18 – 11:22.74 (9.6s)
125
So no, the future isn't about being more human.
11:22.74 – 11:26.26 (3.5s)
126
It's about building the systems that will allow humans to do what matters most.
11:26.30 – 11:31.10 (4.8s)
127
This is not a story about job loss.
11:31.86 – 11:34.26 (2.4s)
128
It is a story about human differentiation.
11:35.22 – 11:37.86 (2.6s)
129
AI will keep on climbing. That is not up to us.
11:39.34 – 11:42.74 (3.4s)
130
But how fast we climb with it, that is up to us.
11:42.74 – 11:46.42 (3.7s)
131
So we need to stop asking: Will there still be jobs for humans?
11:47.30 – 11:52.30 (5.0s)
132
And focus on answering: What do we want humans to be best at?
11:52.34 – 11:57.66 (5.3s)
133
Because in the age of AI, being human isn't a fallback, it's a practice.
11:59.14 – 12:06.54 (7.4s)
134
Let's make it exceptional.
12:06.58 – 12:08.82 (2.2s)
135
Thank you. (Applause)
12:10.14 – 12:16.86 (6.7s)
背景与背景
在这段视频中，演讲者Vinciane Beauchene探讨了人工智能（AI）在未来十年内对工作市场的影响。他反思了阿兰·图灵的图灵测试，并质疑仅仅依靠对话是否能够衡量智能。Beauchene强调，在职场中，人与AI的协作和价值创造才是更为重要的。他着重指出，工作中的关系和人际互动是AI无法完全替代的，这为我们在英语学习中的应用提供了深刻的启示。
日常交流的五个关键短语
- 做什么会改变世界？ - 这个问题促使我们反思人类的重要性。
- 如果AI可以接管所有任务，你会留住谁？ - 提醒我们思考职场中人类的独特价值。
- 建立关系比推销产品更重要。 - 强调在工作中建立联系的重要性。
- 在AI时代，人类的价值并没有消失。 - 传达出人类在变化中的角色。
- 你准备好迎接未来的变化吗？ - 激励我们对未来的适应性保持开放的态度。
逐步影子跟读指南
要有效理解并应用这段演讲的内容，以下是一个英语影子跟读的简单步骤指南，帮助你在观看时进行跟读练习：
- 选择段落：从视频中选择一小段你感兴趣的内容，尤其是含有以上短语的部分。
- 第一遍观看：完整地观看视频，不要尝试跟读，专注于理解演讲者的思想和情感。
- 分段分析：将选取的段落分成小段，逐句播放。每完成一句，暂停视频。
- 跟读练习：试图大声跟读这句话，模仿演讲者的语调和语速。在这个过程中，多次重复，直到流利为止。
- 自我纠正：录下自己的声音，与演讲者的声音进行对比，找出发音、语调的差异并加以纠正。
通过这种看YouTube学英语的方式，不仅提升了你的英语口语能力，同时也能更好地理解人类在职场中的相对价值。英语口语练习的效果会随着你不断的练习而显著提升，让你在面对未来的人工智能挑战时，拥有更强的人际交流能力。
什么是跟读法？
跟读法 (Shadowing) 是一种有科学依据的语言学习技巧，最初开发用于专业口译员的培训，并由多语言者Alexander Arguelles博士普及。这个方法简单而强大：您在听英语母语原声的同时立即大声重复——就像是一个延迟1-2秒紧跟说话者的影子。与被动听力或语法练习不同，跟读法强迫您的大脑和口腔肌肉同时处理并模仿真实的讲话模式。研究表明它能显着提高发音准确性，语调，节奏，连读，听力理解和口语流利度——使其成为雅思口语备考和真实英语交流最有效的方法之一。
如何在ShadowingEnglish上有效练习
- 选择您的视频： 挑选一段语音清晰、自然的YouTube视频。TED演讲，BBC新闻，电影片段，播客或雅思口语范例都很好。将URL粘贴到搜索栏中。从较短的视频（短于5分钟）以及您真正感兴趣的内容开始——兴趣是最重要的导师。
- 先听，理解上下文： 第一次听的时候，将速度保持在1倍速并仅仅倾听。还不要尝试重复。专注于理解其含义，收集新词汇，并注意讲话人如何强调单词，连读声音及使用停顿。
- 设置跟读模式：
- 等待模式：选择
+3s或+5s——在每句话播放完毕后，视频会自动暂停以便您有时间大声重复它。如果您想完全控制并在每次重复后由您自己点击下一步，请选择手动。 - 字幕同步：YouTube字幕有时会在音频前或后略微出现。使用
±100ms使它们完美对齐以助您准确跟读。
- 等待模式：选择
- 大声跟读（核心练习）： 这是真正发生改变的一步。当一个句子播放出来立刻——或在暂停期间——大声、清晰且自信地重复出来。千万不要只是张张嘴：要模仿说话者的准确节奏、重音、音高和连读。力求听上去就像说话者的影子，而不仅是逐字背诵。使用重复功能多次练习同一个句子，直到感觉自然为止。
- 提高难度： 当练习段落变得相对舒适后，就去挑战自我。将速度增加至 <code>1.25x</code> 或甚至 <code>1.5x</code> 以训练高速语言反射。或者将等待模式调整为 <code>关闭</code> 以进行连续跟读——这是最进阶同样收益最大的模式。持续的每日15–30分钟的练习将可以在几周内产生可见的效果。