Three-star electronics made new profits in the second quarter, but the harvest was below market expectations and failed to stabilize the chip plate. On Tuesday, when the semiconductor and storage units in the AI concept were generally weak, the market began to reassess how long the AI transaction, which had been driven by the expansion of the GPU, storage and data centres over time, would last.
The chip unit continues to fall after Tristar performance.
Three stars still fell by nearly 7% that day, exacerbating the sale of AI-related chips. Equity shares such as American Light Technology and Sandisk are also under pressure, reflecting the expected slowdown of investors in terms of high growth in the industry.
Markets are concerned that large cloud service providers may subsequently slow down AI infrastructure inputs. If this trend expands and used to rely on chips and memory blocks supported by high capital expenditure, valuation will face a more direct test.
Funds start to withdraw from existing chip units.
Another Korean storage chip manufacturer, SK Hercules, before he went to the United States this week, the stock price was about 25 percent higher than ever before. The transaction was also seen as a factor in the diversion of market funds, which kept the original AI chip under pressure.
In a high valuation context, investor tolerance of the AI plate is declining. Even if the profits of the enterprise remain high, stock price volatility will significantly increase as long as income, direction or expenditure prospects are not as anticipated.
Open source model drives a new round of revaluation
At the same time, China AI Initials is promoting customized AI chip programmes to support fast-growing open-source GLM model demand. This move reinforces another judgement: the next phase of AI competition may not necessarily be entirely dependent on high-end American chips, but may also shift to lower-cost, more efficient localized software and hardware systems.
This means that the market is revisiting a central issue: Whether future AI development will require a sustained increase in GPU and high bandwidth storage inputs, or whether increased model efficiency will slow down infrastructure needs.
Encryption market or be affected by financial wheeling
The article mentions that the past year has seen AI transactions continue to attract funds and, to some extent, suppress the interest of Bitcoin and the wider encryption market. If investor enthusiasm for the AI plate continues to fall, some of the funds may be redirected to digital assets.
This change also happens after SpaceX's large IPO. As AI stock valuations remain high, discussions in the market on the direction of resource allocation are warming.