算力瓶颈重塑 AI 产业格局

The Economist 2026.05.02 栏目: Leaders

Leaders | Compute says no

人工智能面临算力供给不足的挑战，这将深刻影响其利润分配、发展速度和商业模式。

AI 供应趋紧已成现实

瓶颈环节正在重塑 AI 的经济逻辑

Apr 30th 2026 | 4 min read

Artificial intelligence has a supply problem. As the worldgorges ontokens, thesnippetsof text by which the output of a large language model is counted, it isrunning short ofthem. Weekly token consumption quadrupled between January and March, according to OpenRouter, a marketplace for AI models, partly because of the growing use of coding tools. The industry cannot keep up.

人工智能正面临一个供给问题。随着全世界疯狂消耗 token——也就是用于计算大型语言模型输出量的文本片段——token 正变得供不应求。根据 AI 模型市场 OpenRouter 的数据，1月至3月之间，每周 token 消耗量增长了四倍，部分原因是编程工具的使用不断增加。这个行业已经跟不上需求了。

表达积累:

At model-makers and tech giants,rationing is afoot. Anthropic, maker of Claude, recently adjusted its terms to deter heavy use during peak hours. Amazon says that "capacity constraints" have limited its growth. Sarah Friar, the finance chief of OpenAI, developer of ChatGPT, has said the company is not pursuing every opportunity because it does not have enough processing power (or "compute"). It recentlyscrappedits video-generation model.

在模型开发商和科技巨头那里，配给措施正在出现。Claude 的开发商 Anthropic 最近调整了其服务条款，以抑制高峰时段的重度使用。亚马逊表示，"容量限制"已经制约了其增长。ChatGPT 开发商 OpenAI 的首席财务官 Sarah Friar 曾表示，由于公司没有足够的处理能力，也就是"算力"，所以并没有追求每一个商业机会。OpenAI 最近还取消了其视频生成模型。

表达积累:

The consequences of the supplycrunchcould befar-reaching. A world of scarce compute will shape the economics of AI, changing everything from theallocationof profits to the incentives to use the technology.

这场供给紧缩的后果可能影响深远。一个算力稀缺的世界，将塑造人工智能的经济逻辑，改变从利润分配到技术使用激励机制在内的一切。

表达积累:

Adding AI capacity quickly is hard. Particularly in America, local opposition to new data centres has slowed their construction. Shortages of transformers, switchgear and gas turbines cause delays; some of this equipment can take two to five years to arrive.The tightest bottleneckis in processors. Chips for AI, such as those designed by Nvidia, the world's most valuable company, remain scarce. The squeeze extends to other types of silicon, too, including memory chips and central processing units (CPUs). Few of these constraints willeaseany time soon. Supply chains take years to expand and hardware-makers are still investing more cautiously than the hyperscalers they supply.

快速增加 AI 产能并不容易。尤其是在美国，当地居民对新建数据中心的反对拖慢了建设进度。变压器、开关设备和燃气轮机的短缺也造成延误；其中一些设备可能需要两到五年才能交付。最紧张的瓶颈在处理器方面。用于 AI 的芯片，例如由全球市值最高公司英伟达设计的芯片，仍然供不应求。这种挤压也延伸到了其他类型的硅芯片，包括存储芯片和中央处理器，即 CPU。这些限制因素在短期内几乎都不会缓解。供应链扩张需要多年时间，而硬件制造商的投资仍然比它们所服务的超大规模云厂商更加谨慎。

表达积累:

When hardware is expensive the size of your balance-sheetmattersmore than ever. Whichever part of the supply chain you look at, only a handful of firms have thefinancial muscleand bargaining power to lock up the hardware they need. This year the five data-centre "hyperscalers"—Amazon, Google, Meta, Microsoft and Oracle—will togethershell outmore than $750bn on capital expenditure. OpenAI and Anthropic have announced hundreds of billions of dollars in partnerships and investments. Nvidia is said to have bought most of the memory it will need in 2026 and part of 2027 well in advance. It has also invested across a range of tech firms toshore upits supply chain.

当硬件价格昂贵时，资产负债表的规模就比以往任何时候都更加重要。无论你观察供应链的哪个环节，只有少数公司拥有足够的财力和议价能力，能够锁定它们所需的硬件。今年，五大数据中心"超大规模厂商"——亚马逊、谷歌、Meta、微软和甲骨文——合计将在资本支出上投入超过7500亿美元。OpenAI 和 Anthropic 已经宣布了数千亿美元规模的合作和投资。据说，英伟达已经提前买下了其在2026年所需的大部分存储芯片，以及2027年所需的一部分。它还投资了一系列科技公司，以稳固自己的供应链。

表达积累:

The greatest profits will be found atchoke points. The AI boom has especially benefited Nvidia and TSMC, the Taiwanese manufacturer that makes almost all of the most advanced chips. Chip manufacturers' pricing power has become as enormous as their transistors are tiny. Nvidia's gross margin is about 75%, up from 60% in 2019. TSMC's gross margin is above 60%, roughly twice that of many other contract manufacturers. The hardware giants also haveswayover who gets scarce kit, though they deny picking favourites.

最大的利润将出现在瓶颈环节。AI 热潮尤其让英伟达和台积电受益。台积电是台湾芯片制造商，几乎生产了所有最先进的芯片。芯片制造商的定价权已经变得极其巨大，正如它们的晶体管极其微小一样。英伟达的毛利率约为75%，高于2019年的60%。台积电的毛利率超过60%，大约是许多其他代工制造商的两倍。硬件巨头还对谁能获得稀缺设备拥有影响力，尽管它们否认会偏袒某些客户。

表达积累:

High prices are causing software makers to do more themselves. Custom chips can cost about half as much as buying from Nvidia. But they are not easy to design. Among the software firms doing so, only Google has managed to create a viable alternative in large volumes, and its effort began more than a decade ago. Displacing TSMC is harder still. Other chipmakers, including Intel and Samsung, have struggled to match it at theleading edge. Elon Musk, the boss of SpaceX and Tesla, hasfloateda plan for a "Terafab" to rival TSMC. Its estimated cost is a fantastical $5trn-13trn.

高昂的价格正在促使软件制造商更多地自己动手。定制芯片的成本可能只有购买英伟达芯片的一半左右。但设计定制芯片并不容易。在正在这样做的软件公司中，只有谷歌成功大规模创造出一种可行的替代方案，而它的努力早在十多年前就已经开始。取代台积电则更加困难。包括英特尔和三星在内的其他芯片制造商，一直难以在最先进工艺上追赶台积电。SpaceX 和特斯拉的老板埃隆·马斯克曾提出一个名为"Terafab"的计划，试图与台积电竞争。这个计划的估计成本高得近乎荒诞，达到5万亿至13万亿美元。

表达积累:

The last consequence of the crunch will be to slow theuptakeof the technology. So far the AI boom hasrested onthe cheery assumption that answering queries will only get cheaper. And it has: "inference" prices have fallen by five- to ten-fold in a year. In countries such as India, AI firms are offeringcut-pricesubscriptions to lure users. But that obscures how much cash firms are burning through to sustain those falling prices. OpenAI and Anthropic are expected to lose billions of dollars in the coming years. As both prepare to go public, they will be eager to show that they can one day make a profit.

这场紧缩的最后一个后果，将是减缓这项技术的普及速度。到目前为止，AI 热潮建立在一个乐观假设之上：回答查询的成本只会越来越低。而事实也确实如此："推理"价格在一年内下降了五到十倍。在印度等国家，AI 公司正在提供低价订阅服务，以吸引用户。但这掩盖了一个事实：为了维持这些不断下降的价格，企业正在烧掉大量现金。预计 OpenAI 和 Anthropic 在未来几年将亏损数十亿美元。随着两家公司都准备上市，它们将急于证明自己终有一天能够盈利。

表达积累:

As people find more uses for AI—including applications outside tech, where it is currently most popular—prices will rise. If AI is to transform the economy, demand for tokens willgrow by orders of magnitude. As model-makerspass ontheir rising compute costs, users will have to economise. Today many companies judge themselves by whether or not they use AI at all for a given task. Increasingly, as with human labour, they will have to ask whether they are using AI efficiently. ■

随着人们发现 AI 的更多用途——包括科技行业之外的应用场景，而目前 AI 最受欢迎的地方仍然是科技行业——价格将会上涨。如果 AI 要改变整个经济，对 token 的需求将呈数量级增长。随着模型开发商把不断上升的算力成本转嫁出去，用户将不得不精打细算。如今，许多公司评价自己的标准是：某项任务到底有没有使用 AI。未来，正如对待人类劳动力一样，它们将越来越需要追问：自己是否在高效地使用 AI。■

表达积累: