AI is driving an excessive surge in hardware demand against natural growth patterns

In the consumer hardware industry, there's a term that's hard to avoid recently: price hikes.

In the past couple of days, Apple first raised the prices of its Mac and iPad, and Microsoft quickly followed suit by announcing a price increase for its Xbox. Among the remaining two of the "Big Three" in the game console market, Sony's PS5 and Nintendo's Switch 2 both raised their prices this year. In the domestic market, major mobile phone manufacturers such as Xiaomi, OPPO, vivo, and Honor have also adjusted their prices.

Over the past decade or so, in the highly competitive consumer market, consumers have gotten used to getting larger memory, more storage, and more powerful chips for the same amount of money when upgrading. Now, the situation has reversed. Manufacturers either reduce the configuration or choose to raise the prices.

The cause of this round of price hikes is no longer unfamiliar to the public - a hardware shockwave triggered by AI.

Some overseas reports have started to refer to the price hikes by Apple and Microsoft as the "AI price shock" or "chipflation." The tech media The Verge was even more direct, calling this memory price hike the "RAMageddon."

Put simply: while ordinary people are still anxious about being replaced by AI at work, they find that AI has made computers, mobile phones, and game consoles more expensive.

Behind this hardware winter, it's not just a round of supply - chain fluctuations.

AI data centers are snatching up memory and storage production capacity. Memory manufacturers are shifting more resources to dynamic random - access memory (DRAM), high - bandwidth memory (HBM), and enterprise - level solid - state drives (SSD) for servers. Consumer electronics manufacturers have to accept this round of supply - chain turmoil.

The information technology research and consulting firm Gartner predicted in February this year that by the end of 2026, the combined price of DRAM and SSD will increase by 130%, pushing up the price of PCs by 17% and the price of smartphones by 13%. The global PC shipment volume is expected to decline by 10.4%, and the smartphone shipment volume is expected to decline by 8.4%.

TrendForce's short - term price increase forecast is even more eye - catching. In the second quarter of 2026, the contract price of traditional DRAM is expected to increase by 58% to 63% quarter - on - quarter, and the contract price of NAND Flash is expected to increase by 70% to 75% quarter - on - quarter.

These two sets of data show that the price hikes are not just a short - term pricing move by a certain brand. The position of consumer hardware manufacturers in the supply chain is changing. In the past, mobile phone and PC manufacturers waited for the memory prices to drop and created generations of "well - rounded phones." Now, AI data centers lock in upstream production capacity with larger orders and longer contracts, and terminal manufacturers have become the ones queuing up to pay the bill.

Behind the supply - chain imbalance caused by the memory price hikes, AI is rewriting the situation of consumer hardware companies. Terminal manufacturers are promoting AI mobile phones and AI PCs (artificial intelligence personal computers) while facing more expensive memory, higher storage costs, and more cautious consumers.

Three years ago, in OpenAI, which triggered the AI wave, CEO Altman brought large - scale models to the public's attention. Three years later, this "AI shockwave" that started with Altman has passed through data centers, memory manufacturers, and the terminal supply chain, sweeping across the entire consumer hardware industry.

A number of technology hardware companies that once led the industry in market value are now forced to stand in the aftermath of this round of AI impact.

01

Signs of the impact of this round of AI shockwave on the mobile phone industry emerged last year.

After the release of Xiaomi's Redmi K90 series in October last year, the price controversy quickly made headlines. Public information shows that the base version of the Redmi K90 with 12GB of memory and 256GB of storage is priced at 2,599 yuan, 100 yuan higher than the initial launch price of the previous - generation K80 base version.

Lu Weibing, the president of Xiaomi Group, later explained on social platforms that the cost increase of memory chips "far exceeded expectations" and had been passed on to the pricing of new products. A few days later, Xiaomi reduced the price of the 12GB memory and 512GB storage version by 300 yuan in the first month, bringing it down to 2,899 yuan.

A pricing fluctuation caused by a single mobile phone might just be a product strategy adjustment for other companies. But when it happened to Xiaomi, one of the leading domestic mobile phone brands, it was more like a crack that affected the entire industry.

Over the past decade or so, domestic mobile phone brands represented by Xiaomi have always aimed at "well - rounded phones" with full - fledged configurations. Large memory, large storage, high - refresh - rate screens, and fast - charging technology, which originally belonged to flagship phones, have been gradually introduced to mid - range phones.

Domestic mobile phone users have also become the beneficiaries of the "highly competitive" market: they can get more configurations for the same amount of money.

12GB and 16GB of running memory, 256GB and 512GB of storage - these configurations have been gradually introduced to the mid - range phone price segment from flagship phones in the past few years. Ordinary users can get a well - rounded phone with just over 2,000 yuan.

However, the premise for this strategy to work is the stable supply and market conditions of memory and storage.

By November last year, this series of cost pressures had become the main topic of Xiaomi's earnings conference.

Lu Weibing said after the earnings conference that the global memory prices had risen due to the AI boom, which was pushing up the cost of smartphones. Xiaomi shipped 43.3 million smartphones in that quarter, only increasing by 0.5% year - on - year. The group's revenue increased by 22.3%, mainly from electric vehicles and AI - related businesses.

In other words, the mobile phone business is still Xiaomi's core business, but Xiaomi's new growth pressure has shifted more towards the automotive, Internet of Things (IoT), and AI - related businesses.

In 2026, this pressure can be seen in the profit statements of listed companies. In the latest financial report, Xiaomi's net profit in the first quarter dropped by 43%, and the mobile phone business was suppressed by the high memory cost.

Lu Weibing called the high memory price the "new normal" of the industry after the earnings conference and also predicted that the cost increase would narrow from the third quarter.

Xiaomi is just the first to expose this pressure. Behind it is the price mechanism that domestic mobile phones have relied on for the past decade or so: a mature supply chain, the introduction of high - end configurations to lower - end products, and cost reduction through scale.

When the prices of memory and storage were steadily dropping, manufacturers could put larger running memory and more storage into cheaper devices. Once the prices of these two components rise, the entire mobile phone price segment will be re - adjusted.

In fact, not only Xiaomi, but the entire mobile phone camp is suffering from the impact of this AI shockwave.

Two months ago, vivo raised the prices of its mobile phones due to the increase in semiconductor and memory costs, joining the ranks of manufacturers such as OPPO and Honor. The well - known blogger "Xiaobai CePing" statistics show that the price increase of many recent mobile phone products ranges from 200 yuan to 500 yuan.

Public information shows that the memory cost of a mobile phone with 12GB of memory and 256GB of storage has increased by about 619 yuan.

If high - end phones can still maintain their price bottom line with their profit margins, low - end phones, which support the sales volume, have become the primary victims of the AI shockwave.

The market research firm IDC's forecast shows that the global smartphone market shipment volume will decline by 12.9% in 2026, dropping to 1.12 billion units, the largest decline in history.

IDC's judgment in this forecast is that smartphones below $100 may have difficulty regaining commercial viability even after the memory prices stabilize, because these products rely on extremely low material costs and extremely thin profit margins.

The market research firm Counterpoint Research's forecast last month was even more pessimistic: the global smartphone shipment volume will decline by 13.9% in 2026, dropping to 1.08 billion units, the lowest level since 2013. The low - end and mid - low - end markets are under the most pressure.

Carl Pei, the founder of the mobile phone brand Nothing, publicly talked about the pressure on low - price smartphones at the beginning of this year. He said that in the past, mobile phone companies could give users better configurations every year while maintaining similar prices. Now, the price of some memory modules has risen from less than $20 to more than $100.

Meanwhile, the impact of this AI shockwave is far from reaching its turning point, and the pressure is not limited to low - end phones.

In April this year, at the Huawei Pura series press conference, Yu Chengdong said when talking about the memory price hike, "The pricing pressure for new phones is very high, and the cost has increased a lot. If we can't bear it later, we may also raise the prices."

Subsequently, many media reports revealed that the unit cost of the Huawei Pura series has increased by about 1,200 yuan to 1,500 yuan due to the price hikes in the supply chain such as memory.

In other words, the stronger the brand, the more space it has to pass on the cost; the lower the price segment, the less room it has to bear the cost.

This week, Apple and Microsoft, two giants in the consumer hardware market, have also joined the price - hike camp: Mac, iPad, and Xbox are all on the price - increase list.

It's worth noting that Apple has not yet chosen to raise the price of the iPhone. However, considering that a new generation of iPhone will be launched in the fall, it may just be a matter of time before Apple raises the prices of all its products.

Behind the short - term cash - flow pressure is the change in the game rules. Terminal manufacturers have become tenants in the memory industry. Under the pressure of high rents, if they don't want to quit the game, they can only withstand this round of AI shockwave and keep going.

02

The price hikes of mobile phones are just the surface phenomenon of this AI shockwave.

Deeper changes are taking place inside the hardware industry chain, especially in the upstream supply chain.

On June 25th, Micron's stock price soared by 18.4%, and its market value rose to about $1.4 trillion, briefly surpassing Meta and approaching Tesla. Micron's latest quarterly revenue increased from $9.3 billion a year ago to $41.5 billion, and its net profit increased from $1.9 billion to $28.2 billion.

Micron CEO Sanjay Mehrotra judged at the earnings conference that the memory supply shortage may last until 2027 and will gradually ease until 2028.

In the past, the memory industry often fluctuated greatly between price hikes and price cuts. Mobile phone and PC manufacturers were also used to waiting for the prices to drop before putting larger memory and more storage into the next - generation products.

Now, AI data centers have interrupted this rhythm. Cloud providers and AI companies are no longer just buying goods on a quarterly basis but are locking in the supply for the next few years in advance.

Public information shows that Micron, Samsung, and SK Hynix are promoting long - term supply agreements for AI customers, hoping to get rid of the long - standing cyclical fluctuations in the memory industry. Micron has locked in supply commitments of about $22 billion for several years from customers such as Nvidia, and some agreements include prepayments, price floors, and "take - or - pay" clauses.

In other words, memory is changing from a consumer electronics cyclical product to a strategic component for AI data centers.

The change in HBM is more obvious. AI chips require memory with higher bandwidth and lower latency, and HBM has become an important supporting component for AI accelerator cards from companies such as Nvidia and AMD.

In the past few years, SK Hynix has become one of Nvidia's core suppliers thanks to its HBM supply advantage and has surpassed Samsung in market value to become the highest - valued listed company in South Korea.

This presents a reality for the consumer hardware industry: for the same memory, AI server customers get long - term supply guarantees, while mobile phone, PC, and game console manufacturers mostly receive price - increase notices.

Server hardware is the second type of beneficiary.

If memory manufacturers are selling the key materials needed for AI expansion, Dell and Lenovo are selling the delivery capabilities of AI data centers. These PC giants, which were originally more affected by the memory price hikes, are now also looking for new growth curves through AI infrastructure.

Dell's change is particularly obvious. In May this year, Dell raised its revenue forecast for AI servers in fiscal year 2027 from $50 billion to $60 billion, and at the same time, raised its annual revenue forecast from $138 billion to $142 billion to $165 billion to $169 billion. In the following quarter, Dell's AI server revenue reached $16.1 billion, exceeding the $14.6 billion of its PC business.

However, AI servers are not an easy business either.

Dell COO Jeff Clarke said in the earnings conference call that the company is "re - pricing almost every day," and customers can also feel this pain.

It's not just Dell that has this feeling. Lenovo is also in a similar situation.

In February this year, Lenovo CEO Yang Yuanqing said after the earnings that the memory shortage would put pressure on PC shipments. Lenovo would offset part of the memory cost through price hikes and at the same time accelerate its entry into the AI inference market. By May, Lenovo disclosed that the revenue of its infrastructure solutions business in the fourth fiscal quarter increased by 37%, and the order pipeline for AI servers reached $21 billion.

That is to say, AI has not destroyed all "hardware" businesses. Data center hardware closely related to the AI industry, including memory, servers, and storage, is continuously benefiting.

Although consumer - grade hardware is also hardware, it is not at the core of the current AI industry and bears the price pressure after the cost spill - over.

If a company is not at the core of the AI industry chain, it is more likely to be forced to bear the price - hike pressure brought by AI. Even Apple and Sony are no exception.

Although they are all in the hardware business, their fates are different. So, how can consumer hardware companies catch the AI express train?

Obviously, hardware manufacturers will not sit idle. The concept of AI empowerment has penetrated into the product level, and this question will be answered by AI PCs and AI mobile phones.

03

Recently, with the concentrated outbreak of AI productivity demand and the Agent ecosystem, AI PCs are becoming an important growth pole in the PC industry.

After the post - pandemic replacement dividend faded, it was difficult for traditional PCs to tell a new growth story with features such as screen, portability, and battery life. After the emergence of AI PCs, manufacturers finally have a new selling point: an enterprise office assistant composed of local AI and offline capabilities.

This time, PCs have become the "next - generation productivity entry point."

Gartner predicts that the shipment volume of AI PCs will reach 143 million units in 2026, accounting for 55% of the global PC market. By 2029, AI PCs will become the norm in the PC market.

Take HP as an example. This PC manufacturer's second - quarter revenue increased by