Woofun AI reports that Nobel laureate Paul Krugman has issued a stark warning regarding the stability of the global artificial intelligence sector, characterizing the current market environment as a 'quasi-bubble' teetering on the edge of collapse. This assessment follows a period of intense volatility where the semiconductor industries in both the United States and South Korea faced severe adjustments, prompting Krugman to highlight the critical timing of these declines amidst a broader cooling of industry enthusiasm. The economist's commentary arrives shortly after he publicly criticized Elon Musk, labeling the valuation mechanics surrounding Musk's enterprises as a 'real-life Ponzi scheme' driven by a 'genius image' rather than rational commercial assessment. Krugman argues that the absurd valuations assigned to entities like SpaceX rely entirely on the expectation of Musk's 'genius aura,' creating a feedback loop where rising market caps reinforce the very reputation that inflates them, a dynamic he views as fundamentally unsustainable.

On June 24th, Beijing time, Krugman released a video titled 'The Chips Are Down' to dissect the specific mechanics behind the recent market turbulence. He pointed to the events of Tuesday, June 23rd, local time, as the catalyst for a significant correction across major indices. During this session, the KOSPI index in South Korea plummeted by 9.99%, dragging major semiconductor manufacturers Samsung Electronics and SK Hynix into substantial losses. Simultaneously, the U.S. market experienced a downturn where the Nasdaq index dropped by 1.32%, while the Philadelphia Semiconductor Index suffered a much steeper decline of 7.87%. Krugman posits that attempting to construct a precise logical narrative for these specific price movements is often futile, yet the underlying driver is clearly a shift in market sentiment regarding the practical application of artificial intelligence. Research reports increasingly indicate a disconnect between the volume of content generated by AI models and the actual economic benefits derived from such outputs, a discrepancy that is particularly acute in code generation and pervasive across various industrial sectors.

A more critical variable is the sudden and dramatic reversal in corporate strategy regarding AI adoption, which has shifted from forced integration to active reduction. Not long ago, numerous organizations mandated extensive AI usage among employees, linking performance evaluations directly to the frequency of tool utilization regardless of practical value or employee willingness. Some firms even established quantitative targets requiring the maximization of AI computing resource consumption.

However, as hash rate resources have become increasingly scarce and chip prices have soared, major AI vendors have begun imposing fees that render the marginal cost of heavy AI usage prohibitively high. Consequently, companies have executed a 180-degree turn, instructing staff to curtail AI usage to lower demand for computing power. This strategic pivot underscores the fragility of the previous growth model, where volume was prioritized over efficiency and cost-effectiveness.

Krugman describes this corporate shift as a microcosm of a larger, unsustainable trend, noting that he intends to publish an article detailing how the current wave of AI enthusiasm lacks internal drivers for genuine industrial growth. He observes that while some individuals adopt AI for novelty or experimentation, the primary motivation for many American corporations is not rooted in authentic business needs. Instead, adoption is driven by a fear of missing out and pressure from capital markets, which demand that companies remain at the forefront of AI development to avoid significant valuation reductions. This logic, Krugman asserts, is fundamentally fragile and represents a herd mentality fueled by collective cognitive biases rather than a traditional asset price bubble. He suggests that this illusion has extended too far, creating a precarious situation where market expectations are detached from economic reality.

Regarding the recent market corrections, Krugman maintains that the severity of the declines may be less catastrophic than initially feared, but the timing remains crucial due to the cooling of industry-wide enthusiasm and lowered expectations. He cites an interview with Microsoft CEO Satya Nadella as a pivotal example of this shifting paradigm, where Nadella explicitly stated that allocating all computing power and financial resources to leading AI companies is not feasible. Instead, the focus should shift toward low-cost, lightweight models, a change that would fundamentally alter the market landscape. The current market reaction reflects a realization that high-compute-intensive AI is not a fundamental requirement for industrial development. Krugman concludes that even if investment in AI is significantly reduced, society may still benefit from increased productivity, though this transition could inevitably lead to job displacement.

As of June 25th, Beijing time, the market data reflected continued volatility, with the U.S. Nasdaq index falling for three consecutive days and the Philadelphia Semiconductor Index declining for two days. In contrast, the KOSPI rebounded on June 24th and 25th following the sharp drop on June 23rd. A key factor influencing the performance of the U.S. semiconductor sector recently has been the earnings report from Micron. The company's latest results for the third quarter of fiscal year 2026, covering March to May, revealed revenue reaching $41.46 billion.

Furthermore, Micron's gross margin soared to 84.9%, a figure that exceeded analysts' expectations. This positive financial performance triggered a pre-market surge in Micron's stock price of more than 16% on Thursday, which subsequently lifted the prices of many other storage chip-related stocks.

Another significant development providing support to the market is SK Hynix's plan to list on the Nasdaq on July 10th. According to a report by Reuters on June 25th, Matt Simpson, a senior market analyst at StoneX, noted that stock traders' confidence can be easily restored when a core theme like artificial intelligence dominates the market. Simpson argued that as long as the global economy does not experience a complete collapse, traders will consistently seek reasons to buy at lower prices, with the positive news from Micron serving as the primary catalyst for the week's market boost. Despite this short-term optimism, analysts remain skeptical about the potential for the AI sector to sustain long-term upward trends, as concerns regarding overly high valuations continue to persist within the investment community.

Nick Tewdall, chief market strategist at ATFX Global in Sydney, offered a nuanced perspective on the situation, stating that Micron's earnings report sends a positive signal and that he expects the market to react strongly to these results.

However, Tewdall expressed uncertainty regarding the duration of this industry optimism, warning that ongoing valuation pressures will continue to dampen market sentiment. The views expressed in this analysis are for reference only and do not constitute investment advice, as investing inherently involves risks and caution is advised. This narrative was originally published on the WeChat public account 'JWView' by author Xue Yufei and is authorized by 36Kr.