If AI Takes All The Jobs, This Is What Society Looks Like（英漢雙語 bilingual）如果人工智能取代了所有工作，那麼社會將會變成這樣

https://www.youtube.com/watch?v=PzRDmok1xgY

Moon 1.48M, 2,706 Comments, starting on Sep 1, 2024.

You have got to know what's happening daily or you will be behind the curve of learning. Highly recommended will be closely investigating the AI storm in the making without hesitition. If not, you will have no future. [Mark Wain 02/17/2025]

*

More recent info. on AI effects on jobs as follows:

Which Jobs Will AI Disrupt Most?

https://www.youtube.com/watch?v=tUhZ5PQQtmw

The AI Daily Brief: 

4,592 views  Premiered Feb 15, 2025  The AI Breakdown

AI is reshaping work, but which jobs are most affected? Anthropic's new Economic Index tracks how AI is used across industries, which professions are adopting it fastest, and where automation is already replacing tasks. Software engineers are leading AI adoption, but some fields remain untouched. Are fears of AI replacing jobs overblown, or is it happening faster than expected?

*

From where the profit of capital's data centers comes? DepSeek R1 has answered as follows:

My original question: But according to Marx, only living labor can create surplus value for capital, cam AI data centers hire living labor for profit?

Okay, let's try to tackle this question about Marx, surplus value, and AI data centers. So, the user is asking whether AI data centers can hire living labor for profit, given that Marx believed only living labor creates surplus value. Hmm.

First, I need to recall Marx's theory of surplus value. From what I remember, Marx argued that under capitalism, workers are paid less than the value they produce. The difference between what they produce and their wages is the surplus value, which is the source of profit for capitalists. Machines and technology (what Marx called "constant capital") don't create new value; they just transfer their own value into the product over time. Only "variable capital," which is the labor power of workers, can create surplus value.

My additional explanation is as follows:

According to Marxist political economics, the formula for the social average rate of profit of capital [r] is: r = s/(v +C) = ( s/v )/(1 + C/v ), where [s] is surplus value, s/v is the infamous rate of surplus value [or exploitation rate S], and [C/v] is called the organic composition of capital σ; that is, r = S/(1 + σ). The expression of the decline in the rate of profit caused by technological progress is:

T↑ → σ= C/v ↑ → C↑v↓ → s↓→ r↓

Where T - scientific and technological progress, σ - organic composition, C - constant capital that partially condenses the value of dead labor, that is, the value of means of production, v - variable capital, that is, the value of wages paid by capital to workers, s - surplus value obtained by capital from exploiting workers' living labor without compensation, and r - social average rate of profit. Note: The longer the labor production time and/or the more efficient the living labor, the greater the surplus value created. Surplus value is the source of profit, tax and interest for the capitalist class. [Mark Wain 2025-02-19]

Now, the question is about AI data centers. These are facilities that use artificial intelligence, which likely involves a lot of automated processes, servers, algorithms, etc. But even if the data centers are highly automated, do they still require living labor? Probably. There are people involved in building the data centers, maintaining the servers, programming the AI, managing operations, etc. So even if the day-to-day operations are automated, there's still human labor involved in various stages.

So, according to Marx's theory, the workers in these data centers would be the ones creating surplus value. Even though the AI itself is a form of constant capital, the labor required to develop, maintain, and oversee the AI systems would be the source of surplus value. The owners of the data centers (the capitalists) would pay wages to these workers, and the surplus value generated from their labor would be the basis for profit.

But wait, what about the role of AI in increasing productivity? If AI makes the workers more efficient, does that affect the rate of surplus value? Marx talked about how machinery can increase productivity, leading to relative surplus value. By reducing the necessary labor time (the time workers spend producing the equivalent of their wages), the surplus labor time increases, even if the working day remains the same. So, even if AI is a form of constant capital, its use could lead to higher surplus value by making labor more productive.

However, there's also the question of whether automation can displace workers, leading to what Marx called the "reserve army of labor," which can depress wages. But even in a highly automated industry, there are still some workers employed, and their labor would be the source of surplus value.

Another angle: the workers involved in creating the AI itself. Software engineers, data scientists, etc., who develop the algorithms and systems used in data centers. Their labor is also part of the process. The value of the AI systems would be derived from the labor put into developing them, and when those systems are used in production, their value is transferred to the products (or services) they help create. But the surplus value would still come from the living labor employed in the overall process.

But maybe the user is thinking, if AI becomes so advanced that it can operate with minimal human intervention, would that mean that the surplus value (and thus profit) diminishes? According to Marx, if constant capital becomes too dominant, the rate of profit tends to fall because less variable capital (labor) is employed, leading to less surplus value. This is part of Marx's theory of the tendency of the rate of profit to fall. So, in a highly automated AI data center with very few workers, the rate of profit might decrease over time. However, capitalists might counteract this by finding new areas to exploit labor or by intensifying work (e.g., making remaining workers work harder or longer).

Also, even if the data center itself is automated, the broader economic context includes other industries where labor is exploited. The profits from the data center might rely on the exploitation of labor in other sectors, such as manufacturing the hardware, mining materials for servers, etc. So, while the data center's direct labor might be minimal, it's part of a larger capitalist system that depends on living labor elsewhere.

But the specific question is about AI data centers hiring living labor for profit. So, even if they hire a small number of highly skilled workers, those workers' labor would generate surplus value. For example, a data center might employ technicians to maintain servers, engineers to optimize AI algorithms, security personnel, etc. The wages paid to these workers would be less than the value they contribute, allowing the owners to extract surplus value.

Moreover, the services provided by the AI data center (like data processing, cloud services, etc.) are sold to other businesses. The revenue from these sales comes from other capitalists who, in turn, extract surplus value from their workers. So, the profit of the AI data center is ultimately connected to the surplus value generated in the broader economy.

In summary, according to Marxist theory, even AI data centers require some living labor, and that labor is the source of surplus value, which allows the owners to make a profit. The automation provided by AI is constant capital that transfers its value but doesn't create new value. However, the dynamics of profit might be affected by the composition of capital (more constant vs. variable), leading to tendencies like the falling rate of profit, but individual capitalists can still profit by exploiting the necessary living labor involved in their operations and through the broader exploitation in the economy.

According to Marx's theory of surplus value, only living labor (variable capital) can create new value and surplus value, while machinery and technology (constant capital) merely transfer their existing value into commodities. Applying this framework to AI data centers, we can analyze their profit potential through the lens of Marxist economics:

Necessity of Living Labor:

AI data centers, despite their automation, rely on living labor at multiple stages:

Development & Maintenance: Engineers, programmers, and technicians design, update, and repair AI systems and infrastructure.

Operations: Workers manage server farms, oversee security, and handle data logistics.

Indirect Labor: Labor in related industries (e.g., hardware manufacturing, resource extraction for servers) contributes to the value of the constant capital (AI systems) used in data centers.

These workers' labor power is exploited: they are paid wages less than the value they produce, generating surplus value for capitalists.

Role of AI as Constant Capital:

AI systems themselves are constant capital. They do not create new value but transfer their value (from the labor embedded in their creation) into services or commodities. However, AI can amplify productivity, enabling:

Relative Surplus Value: By reducing necessary labor time (e.g., automating tasks), AI increases surplus labor time for remaining workers.

Cost Reduction: Cheaper production of services (e.g., cloud computing) can lower the value of labor power (wages), indirectly boosting surplus value.

Profitability and the Falling Rate of Profit:

Marx's theory predicts a tendency for the rate of profit to fall as constant capital (e.g., AI infrastructure) displaces variable capital (labor). While AI data centers may employ fewer workers, their profitability depends on:

Exploitation of High-Skill Labor: Technicians and engineers in data centers may face intensified work or wage suppression.

Global Exploitation: Profits may rely on surplus value extracted from workers in supply chains (e.g., mining, manufacturing) or from clients using AI services to exploit their own labor forces.

Broader Capitalist Dynamics:

Even if a data center is minimally staffed, its profits are embedded in a system where surplus value is extracted elsewhere. For example:

Rentier Practices: Charging fees for AI services to firms that exploit labor.

Monopoly Power: Tech giants may leverage AI to dominate markets, redistributing surplus value from other sectors.

R1's Conclusion:  

Under Marxist analysis, AI data centers can hire living labor for profit, as the workers they employ generate surplus value through their labor. While AI (as constant capital) enhances productivity and transfers value, it does not eliminate the need for variable capital. However, heavy reliance on automation may exacerbate the falling rate of profit, pushing capitalists to intensify exploitation, expand into new labor markets, or rely on systemic extraction of surplus value from interconnected industries. Thus, AI data centers remain profitable within capitalism precisely because they are not fully autonomous from living labor.

*

Capitalism Is Over: What's Next Is Worse

https://www.youtube.com/watch?v=hNblIGVKgks

Wisecrack 739,906 views  Feb 7, 2025

*

What was life like for the Japanese after the economic bubble burst? It's so similar to ours! Shocking!(in Chinese)

https://www.youtube.com/watch?v=YBMZiXIOMyc

Hardcore Liu Da Mar 2, 2025

*

I shall comment on the future of AI and its enormous impact below:

We are in the stage of late capitalism, i.e., the end and replacement of capitalism as a social sytem is near. The bourgeoisie of the world has established and maintained the system for more than five hundred years, starting its embryonic stage in today's Holland, through its breaking-through progresses in industrial productions. Finally, its viability has reached its late stage, hence its continued progresses in production have gone to their opposites due to internal social-economic conflicts.

Although reformists, fake lefitists, accomplices as well as helpers of and for the system have done their utmost to maintain the survivability of the system by resorting to all kind of conceivable progresses including taking AI as a panacea, those progresses are doomed to fail.

Long-term decline of the socially averaged rate of profit will cause the system to collapse as a result of the ensuing increase of the rate of unemployment and the  increase of the class struggle both in frequency and intenesity. In other words, the progress in social productive forces including AI can no longer be controllable by the capitalist system. "The contradiction between advanced productive forces and productive relations is a central concept in Marxist theory, explaining societal transformation through historical materialism."[chat.deepseek.com]

The only way out is to change the system to a new one which can handle progresses in social productive forces such as AI. The new system obviously can neither be private-profit driven nor be controlled, ruled and owned by the bourgeoisie and must be controlled by the united proletariat. This new system period is what Karl Marx called "a political transition period, during which the state can only be the dictatorship of the proletariat." [Mark Wain 03-02-2025]

漢譯:

如果人工智能取代了所有工作，那麼社會將會變成這樣

https://www.youtube.com/watch?v=PzRDmok1xgY 

Moon 1.48M，2,706 條評論，從 2024 年 9 月 1 日開始。

你必須知道每天發生的事情，否則你將落後於學習。強烈建議毫不猶豫地密切調查正在形成的人工智能風暴。否則，你將沒有未來。[Mark Wain 02/17/2025]

*

有關人工智能對工作影響的最新信息如下：

人工智能將對哪些工作產生最大影響？（英文）

https://www.youtube.com/watch?v=tUhZ5PQQtmw

人工智能每日簡報：

4,592 次觀看 2025 年 2 月 15 日首播 人工智能細分

人工智能正在重塑工作，但哪些工作受到的影響最大？Anthropic 的新經濟指數追蹤了人工智能在各個行業的使用情況、哪些職業採用人工智能的速度最快以及哪些領域已經由自動化取代了任務。軟件工程師正在引領人工智能的採用，但有些領域仍未受到影響。人們對人工智能取代工作的擔憂是被誇大了，還是它發生的速度比預期的要快？

*

資本的數據中心利潤從何而來？DepSeek R1 的回答的說：

我最初的問題是：但根據馬克思的說法，只有活勞動才能為資本創造剩餘價值，人工智能數據中心會僱傭活勞動來獲利嗎？

好吧，讓我們試着解決這個關於馬克思、剩餘價值和人工智能數據中心的問題。所以，用戶問的是，鑑於馬克思認為只有活勞動才能創造剩餘價值，人工智能數據中心是否可以僱傭活勞動來獲利。嗯。

首先，我需要回顧一下馬克思的剩餘價值理論。我記得馬克思認為，在資本主義制度下，工人的工資[v]低於他們生產的價值[v+s]。他們生產的東西和工資之間的差額就是剩餘價值[s]，這是資本家的利潤來源。機器和技術（馬克思稱之為“不變資本”© ）不會創造新價值；它們只是隨着時間的推移將自己的價值轉移到產品中。只有“可變資本”[v]，即工人的勞動力，才能創造剩餘價值。

我的補充說明如下：

根據馬克思主義政治經濟學，資本的社會平均利潤率[r]的公式是：r = s/(v +C) = ( s/v )/(1 + C/v ), 其中 [s] 是剩餘價值，s/v 就是臭名昭著的的剩餘價值率 【或剝削率 S】，[C/v] 稱作資本的有機組成 σ；即 r = S/(1 + σ)。由技術進步產生的利潤率下降的表達方式是：

T↑ → σ= C/v ↑ → C↑v↓ → s↓→ r↓

其中 T -科技進步, σ - 有機組成, C - 部分地凝結着死勞動價值在內的不變資本也就是生產資料價值, v - 可變資本也就是資本有償付給工人的工資價值,  s - 資本無償剝削工人的活勞動所得到的剩餘價值, 及 r - 社會平均利潤率。注意：勞動生產時間越長及/或其效率越高的活勞動所創造的剩餘價值越大。剩餘價值是資本家階級用來獲取利潤、交稅、付利息的源頭。[Mark Wain 2025-02-19]

現在，問題是關於人工智能數據中心的。這些設施使用人工智能，可能涉及大量自動化流程、服務器、算法等。但即使數據中心高度自動化，它們是否仍然需要活勞動力？很可能。有人參與數據中心的建設、服務器的維護、人工智能的編程、運營管理等。因此，即使日常運營實現了自動化，各個階段仍然需要人工。

因此，根據馬克思的理論，這些數據中心的工人將創造剩餘價值。即使人工智能本身是一種不變資本，開發、維護和監督人工智能系統所需的勞動力將是剩餘價值的來源。數據中心的所有者（資本家）將向這些工人支付工資，他們勞動產生的剩餘價值將成為利潤的基礎。

但是，人工智能在提高生產力方面的作用又如何呢？如果人工智能使工人更有效率，這會影響剩餘價值率嗎？馬克思談到了機器如何提高生產力，從而產生相對剩餘價值。通過減少必要勞動時間（工人生產相當於其工資的產品所花費的時間），即使工作日保持不變，剩餘勞動時間也會增加。因此，即使人工智能是一種不變資本，它的使用也可以通過提高勞動生產率[也就是每小時生產出來的勞動價值量 - MW]來提高剩餘價值。

然而，還有一個問題，即自動化是否會取代工人，從而導致馬克思所說的“勞動後備軍”，這可能會壓低工資。但即使在高度自動化的行業中，仍有一些工人在工作，他們的勞動將成為剩餘價值的來源。

另一個角度：參與創建人工智能本身的工人。軟件工程師、數據科學家等，他們開發數據中心使用的算法和系統。他們的勞動也是這個過程的一部分。人工智能系統的價值將來自開發它們的勞動，當這些系統用於生產時，它們的價值將轉移到它們幫助創造的產品（或服務）上。但剩餘價值仍然來自整個過程中使用的活勞動。

但也許用戶正在想，如果人工智能變得如此先進，以至於它可以在最少的人為干預下運行，這是否意味着剩餘價值（以及利潤）會減少？根據馬克思的說法，如果不變資本變得過於占主導地位，利潤率就會下降，因為使用的可變資本（勞動力）減少，導致剩餘價值減少。這是馬克思利潤率下降趨勢理論的一部分。因此，在一個高度自動化、工人很少的人工智能數據中心，利潤率可能會隨着時間的推移而下降。然而，資本家可能會通過尋找新的領域來剝削勞動力或加強工作（例如，讓剩餘的工人工作更努力或更長時間）來抵消這種影響。

此外，即使數據中心本身是自動化的，更廣泛的經濟背景也包括其他剝削勞動力的行業。數據中心的利潤可能依賴於其他部門對勞動力的剝削，例如製造硬件、為服務器開採材料等。因此，雖然數據中心的直接勞動力可能很少，但它是更大的資本主義體系的一部分，該體系依賴於其他地方的活勞動力。

但具體問題在於人工智能數據中心僱傭活勞動力以牟利。因此，即使他們僱傭了少數高技能工人，這些工人的勞動也會產生剩餘價值。例如，數據中心可能會僱傭技術人員來維護服務器，僱傭工程師來優化人工智能算法，僱傭安全人員等。支付給這些工人的工資將低於他們貢獻的價值，從而使所有者能夠提取剩餘價值。

此外，人工智能數據中心提供的服務（如數據處理、雲服務等）是出售的資本的數據中心利潤從何而來？DepSeek R1 的回答：

我最初的問題是：但根據馬克思的說法，只有活勞動才能為資本創造剩餘價值，人工智能數據中心會僱傭活勞動來獲利嗎？

好吧，讓我們試着解決這個關於馬克思、剩餘價值和人工智能數據中心的問題。所以，用戶問的是，鑑於馬克思認為只有活勞動才能創造剩餘價值，人工智能數據中心是否可以僱傭活勞動來獲利。嗯。

首先，我需要回顧一下馬克思的剩餘價值理論。我記得馬克思認為，在資本主義制度下，工人的工資低於他們生產的價值。他們生產的東西和工資之間的差額就是剩餘價值，這是資本家的利潤來源。機器和技術（馬克思稱之為“不變資本”）不會創造新價值；它們只是隨着時間的推移將自己的價值轉移到產品中。只有“可變資本”，即工人的勞動力，才能創造剩餘價值。

現在，問題是關於人工智能數據中心的。這些設施使用人工智能，可能涉及大量自動化流程、服務器、算法等。但即使數據中心高度自動化，它們是否仍然需要活勞動力？很可能。有人參與數據中心的建設、服務器的維護、人工智能的編程、運營管理等。因此，即使日常運營實現了自動化，各個階段仍然需要人工。

因此，根據馬克思的理論，這些數據中心的工人將創造剩餘價值。即使人工智能本身是一種不變資本，開發、維護和監督人工智能系統所需的勞動力將是剩餘價值的來源。數據中心的所有者（資本家）將向這些工人支付工資，他們勞動產生的剩餘價值將成為利潤的基礎。

但是，人工智能在提高生產力方面的作用又如何呢？如果人工智能使工人更有效率，這會影響剩餘價值率嗎？馬克思談到了機器如何提高生產力，從而產生相對剩餘價值。通過減少必要勞動時間（工人生產相當於其工資的產品所花費的時間），即使工作日保持不變，剩餘勞動時間也會增加。因此，即使人工智能是一種不變資本，它的使用也可以通過提高勞動生產率來提高剩餘價值。

然而，還有一個問題，即自動化是否會取代工人，從而導致馬克思所說的“勞動後備軍”，這可能會壓低工資。但即使在高度自動化的行業中，仍有一些工人在工作，他們的勞動將成為剩餘價值的來源。

另一個角度：參與創建人工智能本身的工人。軟件工程師、數據科學家等，他們開發數據中心使用的算法和系統。他們的勞動也是這個過程的一部分。人工智能系統的價值將來自開發它們的勞動，當這些系統用於生產時，它們的價值將轉移到它們幫助創造的產品（或服務）上。但剩餘價值仍然來自整個過程中使用的活勞動。

但也許用戶正在想，如果人工智能變得如此先進，以至於它可以在最少的人為干預下運行，這是否意味着剩餘價值（以及利潤）會減少？根據馬克思的說法，如果不變資本變得過於占主導地位，利潤率就會下降，因為使用的可變資本（勞動力）減少，導致剩餘價值減少。這是馬克思利潤率下降趨勢理論的一部分。因此，在一個高度自動化、工人很少的人工智能數據中心，利潤率可能會隨着時間的推移而下降。然而，資本家可能會通過尋找新的領域來剝削勞動力或加強工作（例如，讓剩餘的工人工作更努力或更長時間）來抵消這種影響。

此外，即使數據中心本身是自動化的，更廣泛的經濟背景也包括其他剝削勞動力的行業。數據中心的利潤可能依賴於其他部門對勞動力的剝削，例如製造硬件、為服務器開採材料等。因此，雖然數據中心的直接勞動力可能很少，但它是更大的資本主義體系的一部分，該體系依賴於其他地方的活勞動力。

但具體問題在於人工智能數據中心僱傭活勞動力以牟利。因此，即使他們僱傭了少數高技能工人，這些工人的勞動也會產生剩餘價值。例如，數據中心可能會僱傭技術人員來維護服務器，僱傭工程師來優化人工智能算法，僱傭安全人員等。支付給這些工人的工資將低於他們貢獻的價值，從而使所有者能夠提取剩餘價值。

此外，人工智能數據中心提供的服務（如數據處理、雲服務等）是出售的向其他企業銷售產品。這些銷售收入來自其他資本家，而資本家又從工人身上榨取剩餘價值。因此，人工智能數據中心的利潤最終與更廣泛經濟中產生的剩餘價值有關。

總之，根據馬克思主義理論，即使是人工智能數據中心也需要一些活勞動，而這些勞動是剩餘價值的來源，可以讓所有者獲利。人工智能提供的自動化是不變資本，它轉移了價值，但不會創造新價值。然而，利潤的動態可能會受到資本構成（更多是恆定的而不是可變的）的影響，從而導致利潤率下降等趨勢，但個別資本家仍然可以通過剝削其運營中涉及的必要活勞動以及通過經濟中更廣泛的剝削來獲利。

根據馬克思的剩餘價值理論，只有活勞動（可變資本）才能創造新價值和剩餘價值，而機械和技術（不變資本）只是將其現有價值轉化為商品。將此框架應用於人工智能數據中心，我們可以通過馬克思主義經濟學的視角分析其盈利潛力：

活勞動的必要性：

儘管人工智能數據中心實現了自動化，但在多個階段都依賴於活勞動：

開發和維護：工程師、程序員和技術人員設計、更新和修復人工智能系統和基礎設施。

運營：工人管理服務器場、監督安全並處理數據物流。

間接勞動力：相關行業（例如硬件製造、服務器資源提取）的勞動力為數據中心使用的不變資本（人工智能系統）的價值做出了貢獻。

這些工人的勞動力被剝削：他們獲得的工資低於他們生產的價值，為資本家創造了剩餘價值。

人工智能作為不變資本的作用：

人工智能系統本身就是不變資本。它們不會創造新價值，而是將其價值（從其創造過程中嵌入的勞動）轉移到服務或商品中。然而，人工智能可以提高生產力，從而實現：

相對剩餘價值：通過減少必要的勞動時間（例如，自動化任務），人工智能增加了剩餘工人的剩餘勞動時間。

降低成本：更便宜的服務生產（例如，雲計算）可以降低勞動力的價值（工資），間接提高剩餘價值。

盈利能力和利潤率下降：

馬克思的理論預測，隨着不變資本（例如，人工智能基礎設施）取代可變資本（勞動力），利潤率會下降。雖然人工智能數據中心可能雇用更少的工人，但它們的盈利能力取決於：

高技能勞動力的剝削：數據中心的技術人員和工程師可能面臨工作壓力或工資被壓低。

全球剝削：利潤可能依賴於從供應鏈（例如，採礦、製造業）中的工人或使用人工智能服務剝削自己勞動力的客戶那裡提取的剩餘價值。

更廣泛的資本主義動態：

即使數據中心的人員配備最少，其利潤也嵌入在從其他地方提取剩餘價值的系統中。例如：

食利者實踐：向剝削勞動力的公司收取人工智能服務費用。

壟斷力量：科技巨頭可能會利用人工智能來主導市場，重新分配其他行業的剩餘價值。

R1的結論：

根據馬克思主義的分析，人工智能數據中心可以僱傭活勞動力來獲利，因為他們雇用的工人通過他們的勞動創造了剩餘價值[s]。雖然人工智能（作為不變資本©）提高了生產力[P]並轉移了價值，但它並沒有消除對可變資本[v]的需求。然而，對自動化的過度依賴可能會加劇利潤率[r]的下降，迫使資本家加強剝削，擴展到新的勞動力市場，或依靠系統地從相互關聯的行業中提取剩餘價值[s]。因此，人工智能數據中心在資本主義制度下仍然有利可圖，正是因為它們並非完全獨立於活勞動力[v]。

*

資本主義已結束：接下來會更糟（英文）

https://www.youtube.com/watch?v=hNblIGVKgks

Wisecrack 739,906 次觀看 2025 年 2 月 7 日

*

日本人經濟泡沫破裂後的生活是啥樣？和我們太像了！震撼！

https://www.youtube.com/watch?v=YBMZiXIOMyc

硬核劉大 Mar 2, 2025

下面我將對人工智能的未來及其巨大影響作出評論：

我們正處於晚期資本主義階段，即資本主義作為一種社會制度的終結和替代已經臨近。世界資產階級通過工業生產的突破性進步，從今天的荷蘭開始，建立並維持了五百多年的資本主義制度。最後，它的生存能力已經到了晚期，因此，由於內部的社會與經濟的衝突，其在生產方面的持續進步走向了反面。

儘管改良主義者、偽左派、同謀者以及制度的幫閒們竭盡全力維護制度的生存能力，訴諸各種可以想象的進步，包括把人工智能當作靈丹妙藥，但這些進步註定要失敗。

社會平均利潤率的長期下降將導致制度崩潰，隨之而來的是失業率的上升和階級鬥爭的頻率和強度的增加。換言之，包括人工智能在內的社會生產力的進步已經不再是資本主義制度可以控制的。“先進生產力和生產關係之間的矛盾是馬克思主義理論的核心概念，它通過歷史唯物主義來解釋社會變革。”[chat.deepseek.com]

唯一的出路是將制度改變為一個能夠處理人工智能等社會生產力進步的新制度。新制度顯然既不能被私人利潤驅動，也不能被資產階級控制、統治和擁有，而必須由團結起來的無產階級控制。這個新制度時期就是卡爾·馬克思所說的“政治過渡時期，在此期間國家只能是無產階級專政。”[Mark Wain 03-02-2025]