At the very centre of the scientific historical materialist method developed by Karl Marx is the understanding that the objective foundations of revolution are to be found in the contradiction between the growth of the productive forces and the social relations within which they have developed.

In the course of the past century and half since this foundational conception was first elaborated, this contradiction has erupted in the form of economic crises, wars, the intensification of the class struggle and social revolution, most notably the October 1917 Russian revolution.

The development of artificial intelligence (or more correctly augmented intelligence) AI and the growing concern that it has the potential to set off a major economic and financial crises shows that the contradiction identified by Marx is rapidly coming to the surface once again.

AI contains within it the potential for an enormous advance of the productive force in every area of economic activity, possibly the greatest in human history.

But it is running into a headlong conflict with the system of social relations—the capitalist market and profit system, based on private ownership within which it is encased. This conflict is expressed in the fears that while it will bring about vast increases in productivity, this very development will result in economic and financial crises and social devastation.

These fears are manifested in two basic forms. First, the possibility that the massive investments in AI data centres, running into trillions of dollars, will not generate sufficient revenue to return a profit. The second is that the application of AI to a whole swathe of economic and financial activity will simply wipe out a range of companies, many of which have existed for decades, and lead to a crisis for the financial institutions, particularly private credit, that have backed them.

The initial business model for AI, advanced in the wake of the release of OpenAI’s ChatGPT in November 2022, was that the massive investments by so-called hyperscalers, Amazon, Google, Microsoft, Meta and others, combined with the high-end chips developed by Nivida would create monopolistic dominance of AI, enabling them to secure super-profits and pay off their investments.

But this “happy scenario” has come under question on a number of fronts. There is concern over whether the financial arrangements between the major players, in particular the circularity in which one firm provides money to another to enable it to buy its products, has created a false picture similar to that which accompanied the dot.com bubble at the beginning of the century.

Circular deals can give the appearance of confidence and economic health but in the final analysis profitability will be determined by the take-up of AI outside of these confines.

Concerns that the capital spending on AI is outrunning profit generation were seen in the reaction to an announcement by Amazon earlier this month. With news that its investment in 2026 would be $200 billion, up from the $130 billion last year, its share price fell. Amazon shares are now down 20 percent since their peak last November.

There have also been significant falls in the share price of Microsoft, Nvidia and Meta. As the Australian Financial Review noted in a recent article “suddenly, investors are very sceptical that the huge amounts being invested by the likes of Amazon will deliver returns.”

Aside from the issue of whether sufficient revenue can be generated there is the question of moral depreciation. That is, depreciation not through wear and tear, but because of the development of a superior product.

The history of technology shows that an innovation which enjoys superiority and greater profits at one point, can rapidly lose it. Not so long ago the Blackberry cell phone was all the rage—everyone had to have one because of the innovations it had made. But it was very quickly superseded and has largely passed into history.

Something similar is certainly possible in the development of AI, which is still very much in its infancy, with the result that the massive investments undertaken so far could become “stranded assets.”  But while they could lose their value the debts used to finance them would remain.

There was an indication of this possibility when the major AI firms and investors were jolted by the news at the beginning of 2025 that the Chinese start-up company DeepSeek had developed a chatbot as good as those available but produced at a fraction of the cost and utilising less powerful chips because of the bans imposed by the US.

These issues concern what might be characterised as the “failures” of AI within the framework of the profit system. That is, the consequences of any inability to generate the revenues needed to return a sufficient profit on the massive capital investments that have been made, and the potential for the development of catastrophic economic and financial crises.

But what can be termed the “successes” of AI are no less dangerous for the stability of the profit system and its financial structures. This is because the introduction of AI generates additional profit through cost cutting based on the elimination of labour and the potential wiping out of whole firms whose operations are based on a technology now being superseded by AI.

Events so far this month have seen the beginning of what is set to be a major transformation. The release by the AI firm Anthropic of its Claude Cowork platform, which has the capacity to develop codes, among other things, as well as automating functions such as customer support and legal services, sent a shock wave through the world of software firms.

A JPMorgan index of software stocks fell by 7 percent, bringing the total decline for the year to 18 percent.

This month alone, the market value of two of the world’s largest software companies, Salesforce and ServiceNow, has fallen by a fifth, according to a report in the Financial Times.

There are major financial implications because the software market has been a target of private credit firms. It has been estimated that takeovers of software companies financed by private equity funds have accounted for about 40 percent of their trillions of dollars of deal-making activity in the past decade.

The global investment bank UBS has warned that in the worst-case private credit could see default rates rise to as high as 13 percent in the US by AI generated “aggressive” disruption among corporate borrowers. According to UBS analysts as much as 35 percent of the $1.7 trillion private credit market is exposed to the risk of disruption caused by AI.

This week markets were hit with another AI shock as the Los Angeles-based fintech firm Altruist announced the launch of a tax planning tool in its AI platform Hazel. The shares of financial services companies including Charles Schwab and Morgan Stanley were sold off.

It said the new tool enabled advisers to develop strategies for clients “within minutes” after reading tax returns and payslips. Even allowing for a degree of boosterism, remarks by the Altruist CEO Jason Wenk indicate the direction of developments. He claimed the Hazel platform could replace “any platform in wealth management” and that jobs which previously required entire teams could be done with AI “effectively for $100 a month.”

Karl Marx did not foresee the specific development of AI. But he did point to the objective logic that has brought it about. He noted in the Grundrisse, his “rough draft” for Capital, that capital “calls to life all the powers of science and of nature” and provides thereby the foundation for the development of humanity. But at the same time, it has confined these developments within the framework of a social system based on the exploitation of wage labour and profit.

“Forces of production and social relations—two different sides of the development of the social individual—appear to capital as mere means, and are merely means for it to produce on its limited foundation. In fact, they are the material conditions to blow this foundation sky high.” (Marx Grundrisse p.706.)

This “blowing sky high” will take the form of economic and financial crises. It will bring an eruption of class struggle, the beginnings of which can be seen, in which the working class must fight on the basis of a socialist program.

It will be enormously strengthened ideologically and politically in that struggle because all the objections advanced by the bourgeoisie and its defenders about the “impossibility” of socialism—conscious planning of the economy cannot be undertaken, there is insufficiency of material resources to ensure genuine social equality, it is impossible to liberate human beings from degrading labour, the capitalist system is so complex that nothing other than a system based on the blind destructive force of the market is possible and so on—have been “blown sky high” by the advent of AI.