Apple hits a hard snag in its price squeezing efforts: Chinese chip manufacturers' quotation this time will no longer be made to pander to its whims.

First, clarify the source of this news, then discuss its significance.

According to a report by South Korean media Digital Daily on August 5, 2026, which was reposted by multiple domestic media outlets, Apple Inc., to ease the manufacturing cost pressure of its next-generation products, considered including China's ChangXin Memory Technologies into its supply chain and negotiated on the supply price of mobile DRAM such as LPDDR5X, but its request for price reduction was not accepted. The report states that the quotation insisted on by ChangXin is higher than or at least at a similar level to that of Samsung Electronics and SK Hynix (Source: South Korea's Digital Daily, August 5, 2026, reposted by IT Home, My Drivers, China Fund News, etc.).

It needs to be clarified that this is a report from an overseas media citing industry information. As of press time, neither of the two negotiating parties has made a public confirmation on this. This article does not take the plot of "rejection" as an established fact for inference, but regards it as an entry point for observation. What can truly support the judgment is the following public data that can be verified one by one.

A company listed in July with a market capitalization exceeding 3 trillion yuan

On July 27, 2026, Changxin Technology, the parent company of ChangXin Memory Technologies, was officially listed on the Sci-Tech Innovation Board. Its share price rose by 465.82% on the first day, hitting a peak of 55.03 yuan per share during the trading session, with a total market capitalization of over 3 trillion yuan at close, topping the A-share market capitalization list (Source: public market trading data, July 27, 2026).

The public sorting of production capacity is as follows: As of the beginning of 2026, Changxin has 3 12-inch DRAM wafer fabs in Hefei and Beijing, with a monthly production capacity of about 280,000 to 300,000 wafers, and the capacity utilization rate has remained above 95% for a long time; by the end of 2026, the monthly production capacity is expected to increase to about 350,000 wafers (Source: media sorting based on public information, August 5, 2026; the production capacity data is not the official disclosure of the company, and the original source should be traced back before citation).

On the customer side, according to the prospectus of Changxin Technology, the company has carried out cooperation with core industry customers such as Tencent, Alibaba Cloud, ByteDance, Lenovo, and Xiaomi (Source: Prospectus of Changxin Technology).

The combination of the three facts forms the premise of "daring to say no": the long-term capacity utilization rate of over 95% means that there is no idle production line; the fact that leading domestic customers have locked in orders means there is no shortage of buyers; the market capitalization and financing capacity mean there is no shortage of funds for capacity expansion.

Key data: Changxin Technology's share price rose by 465.82% on the first day of listing on the Sci-Tech Innovation Board on July 27, 2026, with a total closing market capitalization of over 3 trillion yuan (Source: public market data, July 27, 2026); as of the beginning of 2026, the monthly production capacity is about 280,000 to 300,000 wafers, and the capacity utilization rate has been above 95% for a long time (Source: media sorting based on public information, August 5, 2026, the original source needs to be traced back).

The global DRAM game table is being reshuffled by AI

To understand why this negotiation got stuck, you have to look at what happened on the other side of the table.

According to the report of Counterpoint Research, in the second quarter of 2026, Samsung Electronics' revenue share in the global DRAM market increased by 1 percentage point from the first quarter to 39%, Micron's market share reached 25%, only one step away from SK Hynix's 26% (Source: Counterpoint Research, August 2026).

At the same time, Samsung Electronics and SK Hynix are shifting a large amount of production capacity to high value-added AI storage product lines such as HBM4, LPCAMM2, and enterprise-level solid-state drives (Source: report caliber of South Korea's Digital Daily, August 5, 2026).

The ripple effect of this is that the supply of general-purpose DRAM is squeezed. And general-purpose DRAM is exactly the type used in consumer electronics products such as mobile phones, computers, and routers.

With shrinking supply and rigid demand, prices are naturally rising. On the China market side, price increases have reached the end shelves: On the afternoon of August 5, 2026, Richard Yu, Executive Director of Huawei and Chairman of Huawei Terminal BG, stated at Huawei's new product launch that memory prices have risen sharply, and all mobile phones may see large-scale price increases afterwards, otherwise selling at the original price will result in losses (Source: Sina Finance, August 5, 2026).

On the same day, two news pieces point to the same thing: the upstream side has taken the pricing power.

Bargaining power is not shouted out, it is supported by three pillars

To abstract this matter, whether a supplier can say no to its largest buyer depends on three conditions being met at the same time.

The first pillar is the scarcity of production capacity. When the overall industry production capacity is crowded out by high value-added products and general products are in a tight balance, suppliers do not need to exchange orders for price cuts.

The second pillar is customer structure. If a supplier's production capacity has been locked by several long-term contracts, the marginal benefit brought by adding a new large customer may not be greater than the cost of customizing quality certification and cooperating with price cuts for it. The report believes that the key reason why ChangXin can refuse is that Chinese manufacturers such as Huawei and Xiaomi have locked in its DRAM production capacity through long-term contracts (Source: report caliber of South Korea's Digital Daily, August 5, 2026).

The third pillar is the substitution cost. For buyers to switch suppliers, they need to go through the whole process of quality certification, production line adaptation and yield ramp-up, which takes quarters. The longer the cycle, the weaker the buyer's bargaining chips.

Only when the three pillars stand at the same time can the dominance of price negotiations be transferred. Without any one of them, the so-called firm quotation will not last for more than a quarter.

Removable framework · Three pillars of bargaining power: production capacity scarcity × customer structure × substitution cost. In one sentence, you can understand whether a supplier dares to say no to its largest buyer. If any of the three is missing, the tough stance is only a gesture; when all three are available, the pricing power is truly transferred.

This is not a victory or defeat, but a change of position

The following is reasoning for readers' reference. Looking at the long term, the position of China's storage industry has undergone three shifts in the past more than ten years: at the earliest stage, there were no products and it could only rely on full imports; then there were products but it could only act as a price follower, and exchanged the admission ticket for domestic substitution for quotations lower than international manufacturers; now the third state has emerged, where the quotation is equal to or even higher than that of international manufacturers, and orders are not obtained by giving price concessions. From "unavailable to buy" to "available but cheap", and then to "available but not cheap", this is a typical path for supply capacity to transition to pricing capacity. However, whether this state can last depends on two unproven variables: one is how long this AI-driven storage shortage cycle will continue, and the other is whether the long-term contracts of domestic customers will remain stable when prices fall. This paragraph is a logical deduction based on industrial rules, and does not represent the position or plan of any enterprise.

It is necessary to be particularly restrained not to interpret an unconfirmed negotiation rumor as a showdown that has already won or lost.

The storage industry is a strongly cyclical industry. When the cycle goes up, the supplier has pricing power; when the cycle goes down, the production capacity will become a burden again. To judge where China's storage industry has really come, what we look at is not the result of a certain negotiation, but whether its quotation can not fall back to the original point when the next cycle goes down.

What does this have to do with you

The first layer is for people who are preparing to buy mobile phones, computers or assemble their own devices. The upstream quotation will eventually appear on your price tag. When storage prices are at a high level, the capacity and specifications you can buy with the same budget will shrink, which is most obvious in entry-level configurations.

The second layer is practitioners in the semiconductor and consumer electronics industry chains. The shift of pricing power will change the profit distribution of the whole chain: the upstream gross profit rises, the midstream modules face pressure, and the downstream complete machine manufacturers choose between raising prices and reducing configurations. The procurement strategy will also change accordingly, and the value of long-term contracts is significantly amplified in the shortage cycle.

The third layer is all people who pay attention to the division of labor in the manufacturing industry. Storage is just a sample. It illustrates the same thing: after industrial upgrading reaches a certain node, the measurement standard will change from "whether it can be manufactured" to "who determines the price after it is manufactured". This transition point is often more difficult than the technological breakthrough itself, and comes later.

What do you want to say about this? Welcome to share your views in the comment area.

This article is for information sharing and industry analysis only, and does not constitute any investment advice, investment analysis opinion or transaction invitation. The plot of price negotiation in the article comes from the report of South Korea's Digital Daily on August 5, 2026 and the repost by domestic media. As of press time, the relevant enterprises have not made public confirmation, which is not regarded as a factual conclusion; the listing data of Changxin Technology comes from public market transaction information (July 27, 2026), the production capacity data is sorted by media based on public information and the original source needs to be traced back, the market share data comes from Counterpoint Research (August 2026), and Richard Yu's speech comes from Sina Finance report (August 5, 2026). All data shall be subject to the original source. The market is risky and decisions should be made with caution. The content marked as "reasoning" in the text is a logical deduction based on public information and does not represent the official position.

This article is from the WeChat Official Account "BT Business Tech", Author: BT Tech Business, published with authorization from 36Kr.