Author | Jiang Xu Explore More Financial Insights | BT Financial Data Pass The full text comprises 2,689 characters, with an estimated reading time of 9 minutes.

PCs, tablets, and mobile phones form the hardware backbone for AI development. Yet, the most foolish thing AI has done is escalating the prices of digital products.

On August 9, 2026, the National Bureau of Statistics (NBS) released consumer price data for July, revealing a notable uptick: tablet computer prices surged by 11.3% month-on-month, computers by 5.5%, and mobile phones by 1.0%. Collectively, these increases contributed to a roughly 0.03 percentage point rise in the month-on-month Consumer Price Index (CPI). The NBS explicitly attributed this trend to artificial intelligence driving iterative upgrades in consumer electronics, thereby increasing demand and, consequently, prices (Source: National Bureau of Statistics, August 9, 2026).

An 11.3% spike might be misconstrued as 'all tablets uniformly rising in price by 11.3% in a single month.' However, this interpretation is inaccurate. The 'tablet computers' category in the CPI is a consumer classification price index, reflecting the average price change of a representative basket of goods. It is influenced by factors such as new and old model transitions, configuration changes, promotional intensity, and product mix.

What truly merits examination is why this set of July data indicates an upward trend, despite consumers having grown accustomed to 'same configurations getting cheaper' for digital products in recent years. The answer lies beyond the retail level. AI is transforming upstream storage demand, end-device configurations, and new product structures, introducing a new cost chain into the pricing logic of digital products.

1. Not Just One Tablet Model Seeing a Price Hike

Let's first clarify the scope. In July, price indices for tablet computers, computers, and mobile phones rose by 11.3%, 5.5%, and 1.0% month-on-month, respectively (Source: National Bureau of Statistics, August 9, 2026). Year-on-year, computer, tablet, and mobile phone prices increased by 17.4%, 17.2%, and 8.5%, respectively, contributing to a roughly 0.14 percentage point increase in the year-on-year CPI (Source: National Bureau of Statistics, August 9, 2026).

Index increases can stem from various combinations: price hikes for identical products, new models entering the statistical sample, a higher proportion of high-configuration products, the withdrawal of previous promotions, or a combination of these factors. Therefore, seeing 11.3% does not allow for direct inference about transaction price changes for a specific brand or model, nor should it be treated as a shopping price list.

Key Data: In July, tablet computer prices rose 11.3% month-on-month, computers by 5.5%, and mobile phones by 1.0%; year-on-year increases were 17.2%, 17.4%, and 8.5%, respectively (Source: National Bureau of Statistics, August 9, 2026).

2. AI First Altered Configuration Lists

The National Bureau of Statistics links these changes to artificial intelligence-driven iterative upgrades in consumer electronics. This explanation is crucial because AI's impact on hardware pricing often goes beyond simply 'adding a software feature'—it pushes up the baseline configuration of devices.

Consider the following reasoning: when end devices need to handle more local model operations, image processing, real-time translation, or intelligent interaction tasks, manufacturers place greater emphasis on computing power, memory capacity, storage speed, and energy efficiency control. Even if the unit price of individual components does not rise significantly, as long as the average configuration of new products increases, the average price of entire devices may also rise.

This is akin to automobiles upgrading from 'just functional' to 'equipped with more sensors and computing platforms.' Consumers are not just paying more for the same device; the 'standard configuration' within the product basket has changed. For price indices, structural upgrades and individual price hikes are both reflected in the average change.

3. Upstream Supply Tightens

Beyond end-device configuration upgrades, storage supply and demand are also shifting. Micron stated in its Q3 2026 earnings materials that tight supply and demand for DRAM and NAND are expected to extend beyond calendar year 2027, with DRAM industry bit output growth projected in the low-to-mid 20% range for 2026 and NAND growth around 20% (Source: Micron Technology, June 24, 2026).

This set of overseas upstream data does not directly prove that storage chips caused price hikes for a specific Chinese tablet, but it provides important context: AI data centers and end devices are both competing for storage and computing resources, and the upstream sector is no longer as clearly oversupplied as in some previous phases.

Production-side data echoes this trend. In July, ex-factory prices in the computer, communication, and other electronic equipment manufacturing sector rose 4.4% year-on-year (Source: National Bureau of Statistics, August 9, 2026). The Producer Price Index (PPI) does not directly correspond to retail prices or the CPI, but simultaneous upward signals in production and consumption at least indicate that this round of price changes is not confined to the final meter of store shelves.

4. Four Layers of Price Transmission

A Portable Framework · The Four Layers of Digital Product Pricing: Upstream Costs & Supply-Demand → New Product Configuration Upgrades → Model Mix Shifts → Retail Promotion Timing. To analyze digital product prices, first identify which layer is changing, then assess whether those changes will affect your specific model.

The first layer is upstream costs and supply-demand. Prices and availability of key components like storage, advanced chips, and displays determine the cost baseline for device manufacturers. The second layer is configuration upgrades, where new products at the same price point may include more memory, larger storage, and stronger computing power.

The third layer is model mix. When high-end new products concentrate their launches, the weight of expensive products in the statistical basket may increase. The fourth layer is promotion timing. The end of major sales events, changes in subsidies, or the phase-out of older models can all alter the actual transaction prices consumers see compared to listed prices.

Separating these four layers helps explain a seemingly contradictory phenomenon: technological progress typically makes computing power per unit cheaper, yet the total price of a new device may still rise because consumers' requirements for a 'sufficient' configuration also increase simultaneously. Lower costs per unit of performance do not necessarily mean lower overall device bills.

Another judgment cannot be omitted: rising price indices do not equate to 'technological progress becoming ineffective.' The semiconductor and computing industries have long exhibited a trend of declining costs per unit of performance, but consumers purchase entire devices, not a single unit of computing power. As long as new products pack more performance, storage, and features into the same product, unit performance can become cheaper while the average selling price of entire devices may still rise.

This also highlights the difference between AI hardware cycles and past simple replacement cycles. Previously, new generations often differentiated themselves through single-point upgrades like screens, cameras, or appearance. In the AI era, upgrades are more likely to involve multiple components such as processors, memory, storage, cooling, and batteries. Consider the following reasoning: when upgrades shift from 'one component getting stronger' to 'multiple components raising the baseline together,' the cost-saving space achieved by device manufacturers is more easily reabsorbed by added configurations.

Looking further downstream to retail, consumer-perceived prices are also influenced by product tiering. A higher proportion of high-end models pushes up average prices; clearance sales of older models pull some transaction prices down. When these two forces coexist, examining just one or two models on an e-commerce page on a given day makes it difficult to judge where industry prices are headed. The value of the CPI lies precisely in providing a broader observational perspective, but it still requires interpretation in conjunction with configurations, model mixes, and promotional cycles.

5. What This Means for You

The first group consists of people planning to buy new tablets, computers, or mobile phones. An 11.3% figure is useful for gauging market direction but unsuitable for replacing specific model price comparisons. What truly matters is comparing actual transaction prices for products with identical configurations, storage, and generations.

The second group comprises professionals in the consumer electronics supply chain. Price changes must be analyzed through the lenses of the CPI, PPI, and upstream storage supply-demand. Focusing solely on the retail end can easily lead to misinterpreting structural upgrades as mere price hikes.

The third group includes those not planning to upgrade their devices. AI's impact on wallets extends beyond software subscription fees—it is entering the replacement costs of office, learning, and entertainment devices through hardware configuration standards and upstream resource allocation.

What truly warrants remembering from the July data is not that 'tablets suddenly became 11.3% more expensive,' but that the price anchors for consumer electronics are being recalibrated. Technological progress will continue to reduce costs per unit of performance, but when new features simultaneously raise configuration thresholds, the cost savings from cheaper unit performance may not fully translate into lower overall device prices.

What are your thoughts on this? Feel free to share your opinions in the comments section.

Disclaimer: This article is for information sharing and industry analysis only and does not constitute any investment advice, investment analysis opinions, or trading solicitations. Data primarily comes from the National Bureau of Statistics' July 2026 CPI and PPI figures and Micron Technology's Q3 2026 public materials, with original sources taking precedence. Markets carry risks; decisions should be made cautiously. Content marked as 'reasoning' represents logical deductions based on public information and does not represent official positions.

This article is original content from BT Finance and may not be used, reproduced, transmitted, or adapted without permission. Infringement will result in legal liability.