English

AI plunge delivers global shock

The global character of the deepening crisis of the AI boom has been illustrated by the plunge in the Tokyo stock market last Friday and a major selloff in South Korea in the wake of the $3.2 trillion loss in overall market value on Wall Street the previous day.

People look at an electronic display board showing Japan's Nikkei index at a securities firm, Monday, Nov. 9, 2026, in Tokyo. [AP Photo/Eugene Hoshiko]

Further underscoring the global character of the shock is the fact that it was in some ways triggered by the failure of the initial public offering—set to be one of the largest in Australian financial history—of the neocloud company Firmus last month.

The Tokyo market fall was characterised as “carnage,” and followed a small uptick on Wall Street after Thursday’s wipeout.

The tech-heavy Nikkei 225 index and the more broadly based Topix both closed down by nearly 6 percent on the day. According to a report in the Financial Times (FT) traders described trading as “complete carnage.” Since its peak in June, the Nikkei has lost more than 20 percent of its market value. Friday was its worst day since the market plunge of April 2025 that followed US president Trump’s announcement of his global “reciprocal” tariffs.

“A lot of people are being blown out of their risk budgets for their technology positions,” Prashant Bhayani, the chief investment officer for Asia at BNP Wealth Management told the FT. “We’ve gotten used to a very low-volatility environment, but that’s clearly ended.”

Companies connected to AI and chip development were the hardest hit. Share prices rebounded somewhat yesterday but that does not indicate any return to stability.

Shares in Softbank, which has been a major backer of the US giant OpenAI, dropped nearly 11 percent on the day. They are now down almost 60 percent since their peak in June.

The South Korean market was also hard hit with the Kospi index falling by 8.8 percent at one point, triggering a 20-minute suspension of trading in what has become a regular occurrence. The index finished down 6 percent for the day and has fallen 40 percent since its peak in June.

Samsung Electronics and SK Hynix, two of the world’s biggest chipmakers, fell 6.4 percent and 10.4 percent respectively.

Japanese prime minister Sanae Takaichi said the government was “watching the situation closely” and it was hard to tell whether the selloff was temporary or whether it “could persist.”

Most market observers think the latter. Commenting on the Wall Street selloff, one analyst told the FT: “I think the market is going down from here. It can rally this week, next week… but the market is going down further.”

The selloff took place in all the important Asian markets and stocks. The Taiex index in Taiwan was down 4.5 percent for the day, its biggest percentage loss since July. Shares in the Taiwan Semi-Conductor Manufacturing Company, one of the world’s major manufacturers of chips, fell 4.7 percent in Taipei. And as the Wall Street Journal noted “even safe havens like Singapore weren’t spared” with the major market index down by 2.9 percent.

Meanwhile more information is coming to light about the Wall Street selloff last Thursday which came as a major shock. After a brief downturn immediately after the sweeping repudiation of Trump in the mid-term elections there was a significant rise.

In a note on Wednesday the major financial firm Citi issued an upbeat note.

“While the risks surrounding both AI and AI financing are likely to remain with us, we think investors have de-risked sufficiently and equities can move higher,” it said.

The bloodbath the next day demonstrated it could not have been more wrong.

As the FT noted, the dominant narrative that the market could power ahead on the back of the AI boom after weeks of warning signals from credit markets it “spectacularly combusted” in the $3.2 trillion rout.

“The selloff was fuelled,” it said, by a “brutal unwind of leveraged tech positions that had built up among hedge funds—with some hacking back their exposure at the fastest rate since Covid-19 and an exodus of retail investors.”

The fears that the victory of the Democrats and opposition to the buildout of data centres would bring a halt to AI development was a catalyst, not the cause of the rout.

“It’s like a chemical reaction,” a senior equities banker told the FT. “You have one, two, three elements, and then there’s an explosion, The election was the explosive element.”

The fear of the markets is not of the Democrats as such—they are bought and paid for servants of Wall Street. It is what the movement to the left in the population, expressed in the election results, portends. There is nothing which strikes more fear into the hearts of finance capital and its concerns for the future of its vast speculative wealth than the development of the class struggle.

The explosive force of the “chemical reaction” is being magnified by the circular financing which lies at the centre of AI development. This is the process by which AI corporations and AI-related firms buy each other’s products and finance each other either by providing equity or debt.

This was the subject of a Bank for International Settlements analysis issued at the beginning of last month. It noted that investment in AI has reached an “extraordinary scale” and that a “meaningful share of this capital flows through firms that serve simultaneously as suppliers of inputs and investors in firms that buy those inputs.”

It found that between 2021 and 2025 nearly 29 percent of all AI firms’ targets for investment were other AI firms and that 55.2 percent of all AI firms’ incoming investment came from other AI firms. Its conclusion from this and other analysis, including the sharing of common supply chains, was that “a significant proportion and commercial relationships in AI are self-referencing.”

It warned that the dynamics of circular financing could “play out if AI revenue growth and end user demand falls short of firms’ expectations” and that “this could drive significant financial market volatility.”

There were further risks arising from the growing use of private credit and special purpose vehicles—entities that are set up off balance sheet—to finance AI infrastructure and which “can create hidden leverage and interconnected exposures that amplify financial stress during downturns.”

The global selloff last week underscores that these risks are not confined to some indefinite future but are very much present in the current situation.

Loading