a16z New Article: From Crypto Mining to AI Cloud, Why Does 'New Cloud' Burn More Cash as It Grows?

Original Title: Charts of the Week: Head In The Neoclouds
Original Author: Moses Sternstein, a16z

Editor’s Note: Against the backdrop of a new wave of computing power investment driven by generative AI, discussions in the market about AI infrastructure are shifting from "Is there enough GPU?" to "Who can provide computing power sustainably?" As the consensus has emerged that model training, inference demands, and data center expansion are all on the rise, a more fundamental question begins to surface: Can the rapid growth in computing power demand truly translate into stable profits and cash flow?

In the latest release of a16z New Media's "Charts of the Week," author Moses Sternstein delves into new cloud companies like CoreWeave, Nebius, and Applied Digital, discussing the growth, valuation, and profitability contradictions in the AI computing power market, and further extending to horizontal SaaS, model routing, and talent competition in cutting-edge laboratories.

In this article, the author does not simply judge whether AI demand is strong; instead, he dissects the current AI transactions into a set of more fundamental structural issues: how existing infrastructure is being repriced, why revenue growth has not synchronized with market expectations, and why the competitive focus in the AI industry is shifting from mere expansion to efficiency and returns.

First, there is the rediscovery of infrastructure value. In the past, land along railroads, natural gas pipelines, and cable television networks served specific industries, only to be transformed into telecommunications and internet infrastructure later. Today, a similar asset revaluation is occurring again. Some new cloud companies that originally served cryptocurrency mining have operational experience in electricity, data centers, cooling systems, and high-density computing; after the explosion of AI demand, these capabilities quickly transformed into scarce computing power supply. The significance lies in the fact that competition in AI infrastructure does not start from scratch; early advantages often come from the recombination of old assets, energy resources, and engineering capabilities.

Second, there is the coexistence of high revenue growth and profitability uncertainty. The early revenue growth rates of new cloud companies like CoreWeave once exceeded those of cloud giants like AWS at their inception, but the capital markets did not grant them the same level of recognition. The reason is that new clouds are not typical light-asset software businesses. GPU procurement, electricity access, data center construction, chip depreciation, and debt interest will rise in tandem with scale, even faster than revenue growth. This means that revenue expansion can only prove that AI computing power demand is strong, but it does not automatically prove that the business model has a sufficiently high capital return rate. What the market is truly waiting for is whether these companies can convert orders and revenue into sustainable free cash flow.

Third, the value of software is being re-differentiated according to AI impact. In the past, the market worried that generative AI would generally weaken the moats of SaaS companies, but Atlassian's performance shows that AI may also become a tool to increase customer spending and product stickiness. Meanwhile, cybersecurity and observability software continue to receive valuation premiums because AI has expanded potential risks and increased enterprises' reliance on mature solutions. This means that the so-called "SaaS apocalypse" will not occur uniformly. Whether AI is a substitute product, drives down prices, or expands demand is becoming a new standard for software valuation differentiation.

Fourth, AI applications are shifting from "stacking tokens" to optimizing tokens. In the past, companies often tended to directly call the most powerful models or give engineering teams a budget to experiment; now, companies like Databricks are starting to use intelligent routing to match different prices and performance models based on task difficulty, reducing costs while maintaining effectiveness. A decrease in token unit price does not necessarily mean a contraction in total AI spending: as unit costs decrease and application scenarios increase, the total consumption of tokens and overall market size may still continue to rise. Efficiency and demand are not mutually exclusive but may form a mutually reinforcing cycle.

If this article were to be compressed into one judgment, it would be: AI infrastructure has proven it can create rapid growth, but the next phase of victory or defeat will depend on whether companies can convert growth into higher capital efficiency. In this sense, the subjects discussed in this article are no longer just whether CoreWeave can become the next generation of cloud giants, but whether the entire AI industry can transition from computing power expansion to sustainable business returns.

The following is the original content:

Under the Title of 'New Cloud'

In the early 20th century, the Southern Pacific Railroad Company had a large amount of idle construction rights on cleared land connecting cities and towns across the United States. The scope of the railroad rights of way was much wider than the tracks themselves, leaving many corridors available for development along the route.

Thus, this railroad company laid a communication network along the railway line, naming it the "Southern Pacific Railroad Internal Networking Telephony." By the 1970s, the company began to commercialize this network, opening it up to a wider range of users.

Subsequently, two things happened simultaneously: on one hand, the monopoly pattern of the long-distance telephone market came to an end; on the other hand, fiber optic cables began to become commercially viable. The original communication corridors were transformed into fiber optic lines, and this network later became known by its English abbreviation, "Sprint." The assets that once served the railroads thus became the backbone of the telecommunications revolution.

It was not only railroad companies that transformed existing physical networks into larger-scale commercial technology infrastructures.

In the 1980s, Williams Company repurposed idle natural gas pipelines into fiber optic channels, establishing WilTel. This company was later sold and eventually renamed WorldCom. By the 1990s, the one-way coaxial cables laid for cable television businesses also underwent a massive, costly upgrade, ultimately becoming the infrastructure for Comcast and Charter to provide broadband internet services to consumers.

This leads us to another type of enterprise: they also possess ready-made infrastructure, and these assets are now being significantly transformed and repriced to meet the demands of an emerging technology—this is the "new cloud" (neocloud) companies.

In summary, most new cloud companies originally engaged in energy and computing-intensive cryptocurrency mining, and then the AI wave arrived. Suddenly, those who own electricity usage rights, infrastructure, and experience in building and managing high-intensity computing loads—like CoreWeave, which also includes a large number of GPUs—stand on one of the hottest tracks today.

Of course, this is not a strict comparison. However, if we observe the three largest publicly listed new cloud companies, their revenue growth rates are indeed remarkable.

We can only estimate the early cloud business revenues of ultra-large-scale cloud service providers, but the general trend is already clear: new cloud companies are growing rapidly, and significantly faster than the growth rates of the three major cloud service providers at their inception.

It should be noted that in the entire computing power sales market, new cloud companies are still relatively small participants.

They still have a long way to go to reach the scale of ultra-large-scale cloud service providers.

The revenue generated by ultra-large-scale cloud service providers each quarter is several orders of magnitude higher than that of new cloud companies. However, at the same time, CoreWeave achieved $2.6 billion in revenue in just about 25 quarters, which AWS took until its 40th quarter to reach after its launch. Again, it is emphasized: the growth rate of these companies is indeed very fast.

With such high growth rates and riding the tailwind of the AI industry, investors should theoretically be quite excited. To some extent, this is indeed the case, but the reality is more complex.

Although the recent financial reports of these companies have generally performed well, CoreWeave's stock price has still fallen by about 16% over the past year; only Nebius is relatively close to its previous highs.

Therefore, the overall story is still good, but for the largest new cloud companies, the appeal is clearly weaker.

The recent market performance has been relatively flat, partly because many growth expectations may have already been reflected in valuations.

For capital-intensive enterprises like new clouds, the price-to-sales ratio is not the most suitable valuation metric, but it is still intuitive to illustrate the issue. Smaller, faster-growing Nebius and Applied Digital have valuation premiums far exceeding those of the much larger CoreWeave. CoreWeave's revenue is still doubling, but it is no longer keeping pace with the 400% to 450% growth rates of the top companies.

If there is a real issue with new cloud companies, it is not growth, but long-term profitability. New cloud companies need to continuously invest in chips, electricity, and physical infrastructure to scale up, and these costs are not low:

Using CoreWeave as an example, its revenue growth is indeed impressive, but its capital expenditures are even more astonishing. Other significant costs include chip depreciation—where the depreciation amount has already exceeded half of the revenue—and the rising interest expenses incurred from borrowing to build expensive infrastructure ahead of time.

This article does not intend to judge whether the new cloud company will ultimately succeed or whether its current stock price is reasonable. Beyond the hype surrounding the topic itself, what we really want to illustrate is that the new cloud company perfectly embodies the tug-of-war between bulls and bears in the entire AI trading landscape.

On one hand, they are situated in a vertical market—the computing power market—that far exceeds everyone’s previous expectations and continues to expand, creating a historically rare growth rate; on the other hand, the costs of building such enterprises are also at historical highs, requiring significant investments in fixed infrastructure that depreciates continuously.

Is Horizontal SaaS Making a Comeback?

Here’s a brief update on the ever-changing market landscape of the "SaaS apocalypse." One company that suffered the most during the previous software stock sell-off has performed quite well in the past month.

Over the past 30 trading days, horizontal software companies have ranked among the top performers in the IGV software ETF component stocks—although they have already given back some of their gains since the data collection began.

Overall, the fundamentals of these companies remain robust. Especially Atlassian, which has not declined under the impact of AI as the market previously expected.

This productivity software company achieved a "double beat" in performance and guidance: cloud business revenue grew by 31% year-over-year, and the growth rate of revenue backlog orders was even higher. But perhaps the more critical signal is that AI is becoming a booster for business growth rather than a hindrance. Atlassian stated that its AI assistant Rovo has been widely adopted; at the same time, customers using Rovo are spending at nearly twice the growth rate of non-Rovo users.

This is good news for Atlassian, good news for Rovo, and good news for horizontal SaaS.

However, the overall valuation of horizontal SaaS remains slightly lower than that of other software categories.

With few exceptions, including Atlassian, horizontal SaaS companies generally have expected price-to-sales ratios that are below the levels corresponding to the "growth—valuation multiple" trend line.

To reiterate, horizontal SaaS has only gone through a relatively good "month." It is far from enough to convince the market that the "SaaS apocalypse" has been canceled based solely on one month’s performance.

Of course, if your software business belongs to the cybersecurity or observability fields, that’s another story—"SaaS apocalypse" has never occurred for these companies.

The cybersecurity sector continues to significantly outperform other categories in the IGV software ETF. In this field, AI has become a tailwind: the market generally believes that AI has increased awareness of cybersecurity threats, and no enterprise customer would rely on "vibe coding" to cobble together their own security solutions.

Whether this logic ultimately holds true will, of course, require time to test. But at least for now, the situation of traditional software companies is far from uniform.

Investors are highly focused on whether AI will bring gains or cause erosion for each company, continuously adjusting their original judgments with each new batch of data—which is only natural.

Moving Towards the Efficiency Frontier of Token Investment

The market landscape surrounding model usage, token consumption, and token spending management continues to evolve in various interesting ways.

Take Databricks as an example.

On the questions of "Which model should we use?" and "Which model is the best?" Databricks has not adopted a winner-takes-all approach, nor has it simply given engineers a budget to decide how to spend it. Instead, it posed another question: "What if we develop a solution that automatically assigns the right tasks to the right models?"

Databricks is certainly not the only company doing this, but it has developed a "Smart Router," and the actual results have been quite satisfactory.

Reportedly, Databricks’ router can call upon more powerful and expensive models when necessary while using weaker and cheaper models when conditions allow, thereby "continuously reducing the average task cost by over 30%."

Overall, pursuing the "efficiency frontier" of token spending is hard not to be a good thing. This indicates that demand continues to grow, and application scenarios are not only evolving at the cutting edge of performance but also spreading to models that are not as top-tier. In the initial pessimistic narrative, these suboptimal models were originally thought to be quickly eliminated.

As we mentioned earlier, efficiency improvements will expand the coverage of demand, which is precisely the dynamic the market hopes to see, similar to the Jevons Paradox.

Silicon Data’s Token Price Strength Index shows that overall price strength is declining, especially as lower-priced open models occupy a higher share in the continuously expanding market.

It is necessary to clarify a frequently misunderstood concept: these indices measure the cost intensity of token spending, not absolute dollar amounts. It also depends on the quantity of token consumption and the comprehensive cost of tokens. This means that even if the price per token decreases, the total consumption of tokens and total spending amount may still continue to rise.

What is truly important is that overall demand continues to grow, and the pricing and model selection gradually moving towards the efficiency frontier will only further drive this growth. Notably, "AI demand" or "AI adoption" is not a single, homogeneous concept. There remains a significant gap between heavy users and other users. This clearly indicates that "always using the best model" may suit some enterprises but certainly not all.

Today, the market is rapidly forming more alternative options. Overall, this is a good thing.

According to data from the enterprise spending management platform Ramp, all enterprises are increasing AI spending, but the median spending gap between the top 10% of enterprises and the median enterprise, as well as between the top 10% and the top 1% of enterprises, is extremely vast.

Ramp’s data tends to be more biased towards tech companies, and this should be considered when interpreting it. However, the data shows that enterprises in the top 10% of spending have per capita AI spending approximately 50 times that of median enterprises.

This distribution is likely not coincidental. Enterprises that can release more value from AI spending are probably also the ones investing the most—though not every company fits this pattern, at least a significant portion does.

An analysis by Boston Consulting Group of 107 publicly traded companies found that companies in the top two quintiles of token usage have revenue growth significantly faster than other companies.

The core message here is that token demand and usage efficiency are mutually reinforcing: the more value enterprises gain, the more tokens they consume.

This process certainly involves a repeated trade-off between investment and return, and R&D will always include some upfront costs. But for the vast majority of enterprises, indiscriminately "piling up tokens" has never been an effective strategy.

Therefore, it is clearly a good thing that enterprises will increasingly not need to adopt such practices in the future.

The Talent War in Frontier Labs

New Media recently welcomed two outstanding team members to OpenAI, so we’ll conclude with a few interesting charts to look at the talent recruitment situation in frontier AI labs.

Dario Amodei recently expressed concern that employees are placing money above mission. According to data from Levels.fyi, this concern may not be unfounded.

If we understand the data at face value, the salaries offered to engineers at Anthropic are very high, significantly surpassing those of engineers with similar qualifications at companies like Google and Tesla.

It seems that being a member of the tech team is indeed a good thing.

Additionally, there is another set of data that is quite interesting.

According to data from Live Data Technologies—compiled by Truist Securities—the talent sources for various labs show significant overlap as well as clear differences:

Both companies have recruited a considerable number of talents from large-cap tech firms, but only OpenAI has hired from Nvidia and Tesla, and both instances occurred in 2026.

Databricks, Snowflake (recently), Palantir, and DeepMind are also common talent sources for both companies.

Both labs have also hired a substantial number of employees from Salesforce and Stripe.

However, the overlap seems to end there. Anthropic has recruited many talents from SaaS companies, while OpenAI has hardly done so; OpenAI has hired extensively from consumer internet, platform markets, and ad tech companies, whereas Anthropic has relatively fewer hires in these areas, except for Airbnb, Netflix, and Uber.

As for what these differences mean, I'll leave it to you to interpret.