Name: 全球南方國家視印度為 AI 領導者，但其自身勞動市場恐受衝擊！| The Dip (The Global South looks to India for AI leadership but its own labor market is at risk | The Dip)
Uploaded: 2026-02-20T17:07:18.000Z
Duration: 14 min 5 s

The unchecked use of AI adoption could destabilize India's workforce.

未經審查的 AI 採用可能會讓印度勞動力不穩定。

情態副詞

Out of the roughly 900,000 engineering graduates, the tech sector right now is perhaps only absorbing 120,000.

大約有九十萬名工程畢業生，但科技業目前可能只吸收了十二萬人。

轉承詞

A lot of the tech and service sector jobs are being substituted as a result of the adoption of AI.

許多科技業和服務業的工作，因為 AI 採用而受到取代。

The Indian story has to be quite critical.

So perhaps a country like Germany or the UK or the US could say let's outsource the AI labor impact to India, India and let Indian workers deal with it.

所以像德國、英國或美國這樣的國家，可能會說讓我們把 AI 勞動力衝擊外包給印度，讓印度勞工來處理。

Let's turn our attention to India now and the AI impact Summit there has just finished and it's seen a number of high level announcements, including at least $200 billion.

讓我們把注意力轉向印度，那裡的 AI 影響高峰會剛剛結束，並看到了一些高層級的宣佈，包括至少兩千億美元。

That's two separate hundred billion dollar announcements on investments in data centers, the launch of homegrown AI models and speech systems.

這是兩筆獨立的千億美元宣佈，用於資料中心投資、本土 AI 模型和語音系統的推出。

And it's all part of the Indian economy's bigger push for sovereignty and influence on the global stage.

這一切都是印度經濟爭取主權和全球舞臺影響力更大推動的一部分。

I do feel that India is in a position to lead what would be considered as the wide scale technological adoption of how AI and its application across sectors could be seen in the next particularly, you know, five to 10 years and could set a precedence for many of the other global south countries to, to take that leap forward.

我覺得印度有能力引領未來五到十年內，AI 及其跨產業應用的廣泛技術採用，並為許多其他全球南方國家樹立領頭羊的典範。

Having said that, I am also aware that the unchecked use of AI adoption could destabilize India's workforce.

不過，我也知道未經審查的 AI 採用可能會讓印度勞動力不穩定。

And as we know, the broader implications of the use of unchecked AI across sectors could not affect just what you would consider as semi formal or unorganized informal jobs.

而且正如我們所知，AI 在各產業的廣泛使用，其影響不僅限於你所認為的半正式或非組織化非正式工作。

But what you consider also is routine white collar jobs which have lacked advanced technological adaptation.

但你也得考慮到那些缺乏先進技術適應的例行性白領工作。

And what I have in mind is basically jobs that have been created out of the IT sector.

我心裡想的基本上是 IT 產業所創造出來的工作。

I mean India's post 90s growth story and it's landing on the world economy scale has been all about IT and financial services and some of these sectors.

我的意思是，印度 90 年代後的成長故事，以及它在全球經濟舞臺上的地位，一直以來都是圍繞著 IT 和金融服務，以及其中一些產業。

Now what you're seeing is that out of the roughly 900,000 engineering graduates who come out looking out for fresh jobs, the tech sector right now is perhaps only absorbing 1.2 lakhs, which is roughly to what would be 120,000 people.

現在你看到的是，在約 90 萬名尋找新工作的工程畢業生中，科技業目前可能只吸收了 12 萬人左右。

And that is troubling because what you're seeing is a lot of the tech and service sector jobs are being substituted as a result of the adoption of AI.

這點很令人擔憂，因為你會發現，由於 AI 的採用，許多科技業和服務業的工作正在被取代。

And I think that's where the Indian story has to be quite, quite critical.

我認為這正是印度故事必須非常、非常關鍵的地方。

It should not be seen as a backend space for us and you know, let's say European companies to use wide scale adoption across sectors, substituting, let's say capital for labor.

它不應該被視為我們和歐洲公司等，可以在各產業廣泛採用，用資本取代勞動力的後勤基地。

While it can increase or enhance efficiency, but it can also destabilize employment.

雖然這可以提高效率，但同時也可能破壞就業穩定。

Okay, so let's break this down into two parts, starting with the digital economy that you mentioned.

好的，那我們就抽成兩部分來談，先從你剛剛提到的數位經濟開始。

And then we'll get to a deeper dive on the jobs.

然後我們再深入探討工作的部分。

I mean, you've spoken about the success story and France's president seems a bit jealous of India.

我是說，你提到了這個成功的故事，法國總統似乎有點嫉妒印度。

他在一個 Impact Summit 的場合發表了一場演講。

He said that India has built something that no other country in the world can.

他說，印度建立了一個世界上沒有其他國家能做到的東西。

A digital identity for 1.4 billion people, a payment system that now processes 20 billion transactions every month, a health infrastructure that has issued 500 million digital health IDs.

一個讓 14 億人使用的數位身分，一個現在每個月處理 200 億筆交易的支付系統，以及一個已經發出 5 億張數位健康 ID 的健康基礎設施。

這真是來自 Emmanuel Macron 的高度讚揚。

So how would you describe the state of India's digital economy as it is at the moment?

所以，你該怎麼描述印度數位經濟目前的現況呢？

Well, on the digital identity part, the availability or access of, let's say, cheap available data to the average Indian has been a highlight of the last, probably decade, decade and a half of the Indian story.

嗯，就數位身分這個部分來說，一般印度人能夠取得便宜的資料，這絕對是過去十年，或者說十年半的印度故事中的一大亮點。

And digitization has enabled a wider consumption of information coming out from India.

而數位化也讓更多來自印度的資訊得以被廣泛消費。

Digital platforms, including social media.

You have an expansive use of, let's say, social media platforms like YouTube, where almost 97 to 98% now, those of households having access to smartphones, are consuming a lot of information.

你們廣泛使用，像是 YouTube 這類的社群媒體平臺，現在幾乎有 97% 到 98% 的家庭都能使用智慧型手機，並大量消費資訊。

What it does is it changes a lot about what you perceive of your lived experience to be.

這大大改變了你對自身生活經驗的看法。

There's a lot for aspiration now in terms of pushing for upward mobility.

現在有許多關於追求向上流動的渴望。

But where I think the digital identity story is most intriguing is in the case of financial inclusion.

但我認為數位身分的發展最引人入勝的地方，在於普惠金融的案例。

The use of United Payments Interface, what we call as upi, for making payments, including for those working in the informal economy.

使用統一支付介面，也就是我們說的 UPI，來進行支付，包括那些在非正式經濟中工作的人。

In fact, very few people perhaps know that as India's workforce has 90% of its workforce working in the informal sector, almost my sense is 70 to 80% of those transactions happening in the informal economy, particularly in the urban areas, are now being anchored through the financial Inclusion network of upi.

事實上，很少人知道，印度有 90% 的勞動力在非正式部門工作，在我看來，非正式經濟中的交易，特別是在都市地區，大約有 70% 到 80% 現在都透過 UPI 的普惠金融網絡來進行。

And that's, that's partly explaining that how well the digitization process in financial inclusion, getting access to perhaps public distribution systems.

這部分解釋了金融普惠的數位化進程，以及取得公共分配系統的順利程度。

And it's important for the audience to understand that India just does not offer a big consumption economy.

觀眾要了解，印度不只提供了一個龐大的消費經濟體。

It also offers a wide presence of a demographic which is keen to consume more and it doesn't really care much about whether they're imports or domestically homegrown produced goods.

它也擁有一大群熱衷於消費的年輕人口，而且他們不太在意商品是進口的還是國內自產的。

And in the urban sectors, you see that happening a lot more.

而且在都市地區，你會看到這種情況發生得更多。

And much of this consumption is being shaped by the digital process of having availability to apps.

而這些消費很大一部分是由數位化過程塑造的，也就是可以方便地使用各種應用程式。

你有像 blinket、Zepto 這樣的應用程式。

I mean so many of these have these things like essentials, groceries, consumer goods accessible to at the household level.

我的意思是，有這麼多應用程式提供必需品、雜貨、消費品等，讓家家戶戶都能取得。

I mean that's something which is unprecedented.

我的意思是，這是前所未有的。

There's no country I can think of perhaps, maybe Indonesia to a certain extent which has seen that rapid adaptation of digitization at a household level.

我想不到有哪個國家，也許印尼在某種程度上，能在家庭層級看到如此快速的數位化適應。

I mean, it seems like the perfect environment for AI to slot in then.

我的意思是，這似乎是 AI 能夠順利切入的完美環境。

So let's dig in a bit more about what you were saying earlier.

那我們就來深入聊聊你剛剛說的。

How do you think that the impact of AI, for better or for worse, will affect India in a different way to how it might affect other economies like the US or Europe?

你覺得 AI 的影響，無論好壞，會如何以不同於美國或歐洲等經濟體的方式影響印度呢？

Okay, so I think we have to see this in perspective.

好的，我認為我們得從長遠來看。

As I said earlier, the, the traditional high skilled jobs in India, which was in the organized formal sector, which have been part of India's urban growth story.

就像我剛剛說的，印度傳統的高技術工作，是在有組織的正規部門，這也是印度城市成長故事的一部分。

You know, you talk about Bangalore, cities like Pune, Gurgaon, Hyderabad, these are hubs of IT sector created jobs from the 90s, right.

你知道，你說到班加羅爾、浦那、古爾岡、海德拉巴這些城市，它們是 IT 產業的中心，從 90 年代就創造了許多工作機會，對吧。

And they've also been the growth engines for the Indian economy at large.

而且它們也一直是整個印度經濟的成長引擎。

Now what's happening is the entry level coding, testing, documentation, customer support, back end processing of data.

現在發生的是，入門級的程式編寫、測試、文件編寫、客戶支援、後端資料處理。

These were all the kind of opportunities that fresh young graduates were being absorbed amidst a rapidly expanding higher education system.

這些都是在高等教育體系快速擴張的環境下，應屆年輕畢業生得以被吸收的機會。

Now technically, AI systems can perform these tasks quite efficiently and effectively, right?

技術上來說，AI 系統可以相當有效率地執行這些任務，對吧？

They excel in environments where outcomes could be clearly defined and training data is abundant.

它們在結果可以明確定義且訓練資料充足的環境中表現出色。

But I say at least a conversation on oversight, like you have global projections that generally talk about aggregate productivity gains from AI, at least in the medium term, in the Indian context, to be merely incremental.

但我認為至少要討論一下監管，就像你們有全球預測，普遍談論 AI 在中長期能帶來總體生產力提升，在印度這個情境下，頂多是漸進式的。

I don't see transformative change happening, let's say over a few years.

我看不出有什麼轉變性的改變會在幾年內發生。

There are distributional consequences however, which I think we need to look at now.

不過，卻有分配上的影響，我認為我們現在需要關注。

There are structural asymmetries that are present across routine labor jobs.

在例行性勞動工作上，存在結構性的不對稱。

Since so many jobs are in the informal sector, you'd expect the impact of that to be almost unchecked or undocumented.

由於很多工作都在非正式部門，你會預期其影響幾乎是未經審查或未被記錄的。

We may not know the nature of technological disruption that may be caused in the unorganized informal sector.

我們可能不知道在非組織化非正式部門可能造成的技術顛覆的性質。

Which means that the only way you can check for any evidence of what AI is causing is in the organized formal sector, which is in areas like IT financial services.

這表示，要檢視 AI 造成影響的任何證據，唯一的方法就是在組織化的正式部門，像是 IT 金融服務業。

Now a lot of this has labor force participation rate.

現在，這很大一部分與勞動力參與率有關。

India's biggest concerns have been in job creation.

印度最大的擔憂一直是創造就業機會。

If you look at the last 10 years, India is no longer facing issues of underemployment which it used to before, or disguised unemployment is facing something worse.

如果看看過去十年，印度不再面臨以前的就業不足問題，或是隱藏性失業，而是面臨更糟的情況。

It's facing double digit unemployment amongst youth who are educated.

現在是受過教育的年輕人出現兩位數的失業率。

And the worst impact of that is amongst female urban educated youths.

而這其中最嚴重的影響，是出現在城市裡的年輕女性知識份子。

So when you have a context where job creation is a concern, the presence of labor intensive jobs is a concern.

所以當你面臨創造就業機會是一個考量點時，勞力密集型工作的存在就成了一個問題。

If you expect almost 40 to 50% of the work that is already available to be replaced by the adoption of a particular technological force, you are clearly looking at a wide scale impact which is going to be very difficult to be measured.

如果你預期現有工作中，有高達 40% 到 50% 會被某種特定技術力量的採用所取代，那顯然你會看到一個大規模的影響，而且將很難被衡量。

I want to jump in on just one point that you were saying there though, because it sounds like it's exactly the, the industry, the sector which India has excelled at and which has been the massive contributor to the GDP which is going to be the one that is set to be most impacted by AI.

我想針對你剛才說的一點插話一下，因為聽起來，這正是印度一直以來表現優異、對 GDP 貢獻巨大的產業和領域，卻將會是受 AI 影響最大的。

So it's these kind of AI tools where they plug in easily to where work is already kind standardized and digitized and that's in the formal sector.

所以這類 AI 工具很容易就能融入已經標準化和數位化的工作流程中，而這就存在於正式部門。

And it sounds like also because a lot of countries in Europe, the US as well, the UK have shifted, outsourced, offshored a lot of their back office tasks, let's say to India.

而且聽起來，因為歐洲、美國和英國的許多國家都已經將大量的後勤任務轉移、外包、離岸到印度。

And that's been a massive industry when it comes to it.

而這也成了一個非常龐大的產業。

And yet it sounds like what may happen or one significant risk is that the risk to labor as a result of the implementation of AI is also going to be offshored to India as well.

然而聽起來，可能會發生的其中一個重大風險，就是 AI 導入所造成的勞動力衝擊，也可能會被外包到印度。

And it's going to be Indian workers, skilled workers, workers who are highly qualified, who are going to see the brunt of this impact.

而受到這波衝擊最大的，將會是印度的勞工，是那些技術熟練、高學歷的勞工。

So perhaps a country like Germany or the UK or the US could say let's outsource the AI labor impact to India and let Indian workers deal with it.

所以像德國、英國或美國這樣的國家，可能會說，我們把 AI 衝擊的勞動力問題外包給印度，讓印度勞工來處理。

Well, quite accurately, unfortunately that's, that's what I see happening.

嗯，很不幸地，這確實是我所看到的。

But I think one of the things that as a state or as a government that Indian policy environment has to be aware about is this process of offshoring happens in areas which allows for two things to happen.

但我想，印度作為一個國家或政府，必須意識到這種外包過程發生在允許兩件事情發生的領域。

Wage mobility, which is increasing of wages.

薪資流動性，也就是薪資的提升。

India's labor market has been in the organized sector, deeply compressed.

印度的勞動市場，在組織化部門，薪資一直被嚴重壓縮。

I mean IT sector also gives you a lot of evidence you don't have very high wages across IT sector jobs.

我意思是，IT 產業也提供了很多證據，IT 產業的職位並沒有非常高的薪資。

There is however precedence for certain companies like TCS or Infosys which have done dramatically well as Indian companies to set resolutions for high value addition that is happening.

不過，像 TCS 或 Infosys 這類公司，作為印度企業表現非常出色，為高價值附加的實現設定了先例。

But to what extent these companies are able to compete with the technological development that's happening with the widespread use of AI, let's say in the US or in Germany, I have my doubts.

但這些公司能在多大程度上與美國或德國等地廣泛使用 AI 所帶來的技術發展競爭，我個人是有點懷疑的。

And one of the biggest reasons for this is that you don't have the same quality of human capital development that was happening during the 90s when the absorption of Internet, ICT, information communications technology led to this huge boom, changing what was called as the engineering programs from civil services to, sorry, civil engineering to what you call as computer science engineering.

其中一個最大的原因，就是你們沒有像 90 年代那樣高品質的人力資本發展，當時網際網路、資通訊技術的普及，造就了巨大的榮景，改變了所謂的工程學程，從土木工程轉變為電腦科學工程。

And then you had suddenly this flurry of computer science engineers who could do basic coding, who could work on apps, who could create apps very quickly.

然後突然湧現了大量電腦科學工程師，他們能做基本的程式編寫，能處理應用程式，能快速開發應用程式。

In fact, India's digital development is in terms of app development, coding.

事實上，印度的數位發展主要體現在應用程式開發和程式編寫方面。

But the interesting statistic for people to look at is that almost 60 to 70% of the job creation that has happened in India the last at least seven to eight years has been in the unorganized informal category.

但有趣的是，過去至少七到八年，印度創造的就業機會中，有將近 60% 到 70% 是來自非組織化非正式工作。

Which means that the lack of jobs in organized sector is pushing people towards the informal sector of which you would see very little of any impact.

這意味著組織化部門的就業機會不足，正將人們推向非正式部門，而這方面的影響你們幾乎看不到。

So just to circle back to the main challenge, I think technologies that augment or automate easy tasks risk eroding what you would consider as an already exposed employment base.

所以，回到主要的挑戰，我認為那些能增強或自動化簡單任務的技術，可能會侵蝕掉你們認為已經很脆弱的就業基礎。

You can raise productivity without collapsing participation.

你們可以提高生產力，而不必導致參與率崩潰。

But at the same time, this distinction is not very semantic.

但同時，這個區別並不是那麼的字面意思。