AI Suddenly Got More Expensive.

In March 2026, Tencent Cloud raised prices for some AI models by over 400%. Alibaba Cloud also increased AI service prices for certain models on its BaiLian platform twice within a month.

A friend in content creation did the math: He uses API calls to a large model to batch-generate a week's worth of short video scripts (about 50). Charged by token consumption, the cost was about RMB 15 three months ago. In March, with the shutdown of many AI relay stations and successive price hikes by major cloud providers, the same workload now costs nearly RMB 60.

In early 2024, China's daily token calls were 100 billion. By March 2026, that number had surged to 140 trillion.

A 1,000-fold increase in two years.

What is a token?

Asking QianWen, 'What's the weather like in Shenzhen today?' consumes about 200 tokens (including the question and response). Asking DeepSeek to write a 5,000-word project plan consumes about 8,000-10,000 tokens. Tokens are the basic unit of AI information processing and the billing unit for computing power. Think of them as AI's word counter—the more you use, the more computing power you burn.

So, why did AI suddenly get more expensive? Is it because vendors want to raise prices?

01 The Truth Behind the Price Hikes

J.P. Morgan predicts that by 2030, China's AI inference token consumption will grow about 370-fold from 2025 levels. Goldman Sachs forecasts a 24-fold increase in global token consumption over the same period.

The conclusion is clear: AI's computing power consumption is entering exponential growth.

The reason? How we use AI has changed.

Previously, we often asked AI questions, and with a few dozen tokens, the interaction ended. But now, more people are asking AI to 'do things for me.'

For example, 'Help me plan a 5-day trip, including flights, hotel recommendations, sightseeing routes, and food recommendations.' AI doesn't do this in one step—it requires multiple inferences, tool calls, and step-by-step execution. This is called an Agent task in the industry. An Agent task consumes 100-1,000 times more tokens than a simple Q&A.

From Q&A to execution, computing consumption differs by three orders of magnitude.

On the supply side, leading cloud providers like ByteDance, Alibaba, and Tencent have seen their AI infrastructure spending soar. From 2024 to 2025, spending on core AI chip components doubled from $22 billion to $52 billion. High-bandwidth memory (HBM) now accounts for about two-thirds of GPU material costs, up from roughly 50%. Expensive chips, memory, and electricity have driven up computing costs across the board.

So, price hikes aren't a vendor choice—demand is too strong, and capacity can't keep up.

This resembles mobile communications around 2010. Smartphones exploded in popularity, but base stations were insufficient, making data expensive. Only after large-scale 4G and 5G deployment did data become affordable. AI is now at that same crossroads.

Except, the 'AI base stations' aren't being built by the three major carriers. Instead, China is deploying a brand-new infrastructure: the computing power network.

02 The Black Hole of Computing Power

When people think of enterprises using AI, many assume bosses who don't understand AI demand employees use it to review reports or write plans.

That's short-sighted. The bulk of AI's computing consumption isn't there.

Manufacturing. Companies like Foxconn and CATL use AI for visual quality inspection—cameras capture product images, and AI instantly judges good from defective. Running 24/7, a single production line processes millions of images daily, consuming continuous computing power.

Programmers. Tools like Tongyi Lingma and GitHub Copilot help developers write code and find bugs. Millions of developers in China now use AI-assisted programming daily, with each request consuming computing power.

Content. AI generates copy, video scripts, and product descriptions. Much of the copy behind the short videos you watch daily involves AI.

Healthcare. AI reads CT scans and detects early-stage lung cancer, with accuracy approaching or exceeding human levels in some scenarios.

Research. AI accelerates drug discovery, slashing the timeline from target identification to molecule screening from years to a fraction of the time.

Agriculture. Satellite remote sensing plus AI analysis identifies water-deprived or pest-infested fields, guiding farmers on fertilization and pesticide use.

These scenarios share a common trait: They all consume massive computing power. And they're not niche—manufacturing quality affects every product, medical AI impacts everyone's health, and agricultural AI affects food output.

When computing power runs short, price pressures don't stop at cloud providers. They trickle down the chain to ordinary people.

How to solve this? China's answer: Build a 'high-speed rail network for computing power.'

03 The Weight of a Network

Water networks, new power grids, computing power networks, next-gen communication networks, urban underground pipelines, and logistics networks—these are the 'six networks' explicitly outlined in the 15th Five-Year Plan.

They represent new infrastructure. Like the massive high-speed rail and 4G network construction 20 years ago, the state invests first to lower costs and barriers, benefiting society as a whole.

Why can computing power networks stand alongside power and water grids?

Because computing power has become the 'electricity' of the AI era.

A century ago, factories generated their own power. Then the national grid was built, and factories plugged in—no need for on-site generators.

The computing power network follows the same logic. Today, companies buy GPUs and build data centers, incurring high costs and low utilization. Once the network is built, companies can purchase computing power on demand, like buying electricity—pay for what you use, no need for your own power plant.

How is this network scheduled? The core logic is 'Data Computing in East and West Regions.'

China has planned eight national computing hubs. Four in the east—Beijing-Tianjin-Hebei, Yangtze River Delta, Guangdong-Hong Kong-Macao Greater Bay Area, Chengdu-Chongqing—have concentrated users needing millisecond-level responses. Four in the west—Inner Mongolia, Guizhou, Gansu, Ningxia—offer cheap green power and abundant land.

AI inference tasks requiring real-time responses stay in the east. Non-urgent AI training tasks are scheduled to the west, leveraging cheap green power.

This scheduling isn't theoretical—it's already operational. Zhejiang and Xinjiang use time-zone differences for 'peak-shaving' scheduling, reducing token electricity costs by 18%. Western green data centers achieve a PUE (Power Usage Effectiveness) below 1.15, far below the national average.

In the era of large model inference, the ratio of inference to training computing demand is expected to reach 3:1. This means future computing networks will be designed more for real-time responses—exactly the value of Data Computing in East and West Regions.

04 Computing That Reaches for the Stars

Computing power networks aren't just ground-based.

On May 14, 2025, China launched 12 space computing satellites aboard a Long March 2D rocket from Jiuquan, forming the world's first large-scale space intelligence computing satellite constellation.

These satellites carry 'AI brains'—high-performance smart computers and distributed operating systems. Traditionally, satellites must transmit images to the ground for processing, taking hours. Computing satellites analyze data in orbit, compressing disaster emergency response to seconds.

Early identification of wildfires and floods, precision agriculture, ocean monitoring—satellites can 'compute while flying,' transmitting only key results to the ground, drastically reducing data transmission bottlenecks.

An even bolder concept: building large-scale computing centers in space, turning space into 'offshore data centers.'

By late 2025, Elon Musk announced SpaceX's plan to expand the Starlink V3 constellation and build space data centers in orbit, submitting a formal application to the FCC in early 2026 for a million-satellite network. Jeff Bezos predicted in October 2025 that gigawatt-scale data centers will be built in space within 10-20 years. Theoretically feasible, but technically far from practical. China's current focus is the first step: satellite constellations for computing power.

Ground-based networks address 'insufficient computing power,' while space-based networks tackle the bottleneck of massive satellite data that can't be transmitted or processed quickly. These two paths, though at different stages, share the same direction: making computing power ubiquitous.

05 Trillion-Dollar Investments

Total investment in the computing power network during the 15th Five-Year Plan period is expected to reach RMB 2-2.5 trillion.

In 2026 alone, total investment in the 'six networks' exceeds RMB 7 trillion, with annual computing network investment no less than RMB 400 billion.

As of early 2026, the national computing interconnectivity platform has connected 578 resource pools from 155 enterprises across 31 provinces (autonomous regions, municipalities), including 316 EFLOPS of intelligent computing resources and 720,000 GPU accelerator cards.

Funding sources are clear. In 2026, the central government allocated over RMB 2.55 trillion in policy funds to support the 'six networks,' with over RMB 800 billion in ultra-long-term special bonds available for the computing network. These bonds have 20-50 year terms, low interest rates, and don't increase local hidden debt.

Market-driven funds are also pouring in. ByteDance, Alibaba, Tencent, Baidu, and the three major carriers are all ramping up investments in computing network-related projects.

Signs of inclusivity are emerging. A smart tech company in Nantong successfully claimed RMB 15,000 in computing voucher subsidies, while models like 'computing power banks' and 'computing power supermarkets' are being piloted in multiple regions.

All this points to one goal: driving down AI computing costs to make AI accessible to more people.

06 Silent Inclusivity

What changes will the computing network bring to your life?

First, your cost of using AI will drop.

Today, when you subscribe to Doubao Premium or use QianWen's paid version, the fees reflect computing prices. Once the computing network is built, AI service prices will likely keep falling, just as mobile data evolved from RMB 10/MB to today's unlimited plans.

Second, more people can start AI businesses.

Today, launching an AI app requires buying dozens of GPUs, with upfront costs in the millions. Once the computing network is widespread, 'computing vouchers' and pay-as-you-go models will drastically lower barriers. With a good idea, you won't need to raise funds for equipment first.

Third, more AI capabilities will emerge where you least expect them.

Precision agriculture boosts grain yields, stabilizing food prices. AI healthcare makes early screening universal, lowering medical costs. Smart transportation reduces congestion, shortening commutes. These won't be branded as 'AI,' but you'll notice life getting better.

China's network isn't just solving 'insufficient computing power.' We must also recognize that AI shouldn't be a weapon for big companies—it should be infrastructure, like electricity, accessible to everyone.

Today's 140 trillion tokens may reach 1,400 trillion when the computing high-speed rail is complete. By then, the cost of using AI could be a fraction of today's.