It Took Salesforce 20 Years to Reach $30 Billion in Revenue — Anthropic Did It in 3

Artificial intelligence has already produced plenty of eye-popping numbers. Nvidia (NASDAQ:NVDA | NVDA Price Prediction) crossed a $5 trillion market cap. Microsoft (NASDAQ:MSFT) and Amazon (NASDAQ:AMZN) alone are spending nearly $400 billion on AI infrastructure this year. Data center electricity demand is climbing so fast that utilities are suddenly growth stocks again.

But one figure may be the most startling yet. Anthropic went from an annualized revenue run rate of $87 million to $30 billion in just three years — and is now reportedly approaching a $40 billion run rate. For context, Salesforce (NYSE:CRM), one of the greatest enterprise software success stories ever built, took roughly 20 years to reach the same level.

That raises a huge question for investors: Is AI creating the fastest business expansion cycle Silicon Valley has ever seen, or are markets getting ahead of themselves?

Anthropic’s Growth Numbers Border on Absurd

Let’s start with the scale of what happened. Anthropic was founded in 2021 by former OpenAI researchers led by Dario Amodei. By early 2024, the company reportedly operated at an annualized revenue run rate of just $87 million. Fast forward to April, and that figure had exploded to $30 billion. Weeks later, estimates placed the run rate closer to $40 billion.

It means Anthropic may have added revenue faster than almost any software company in modern history. Here’s how that stacks up against major tech leaders:

Granted, revenue “run rate” is not the same thing as reported annual revenue under GAAP accounting standards. Run rate extrapolates current revenue over a 12-month period. Still, regardless of how you look at it, the acceleration is staggering.

Part of the reason is that AI adoption is happening inside an already mature digital economy. Salesforce spent years convincing businesses to move customer data into the cloud. Anthropic entered a market where cloud infrastructure, APIs, enterprise software subscriptions, and hyperscale computing already existed. The rails for hypersonic growth were already built.

Why AI Revenue Is Scaling Faster Than SaaS Ever Did

Surprisingly, this isn’t just about chatbot subscriptions. Anthropic’s Claude models have become deeply embedded in enterprise workflows. According to Amazon and Google partnership announcements, Claude powers coding assistants, enterprise search tools, legal document analysis, and customer support systems across large corporations.

That matters because enterprise AI spending behaves differently from consumer app spending. Once a company integrates an AI model into its software stack, usage tends to compound. More employees use it. More tasks shift to it. More tokens get consumed. Revenue scales alongside usage.

Let’s look at the infrastructure behind that growth:

• Amazon committed up to $25 billion into Anthropic
• Google committed more than $40 billion
• Claude models now compete directly against offerings from OpenAI and Meta Platforms (NASDAQ:META)
• Enterprise AI spending is projected to hit up to $1.7 trillion annually by 2032, according to Fortune Business Insights

In any case, investors should understand this growth comes with enormous costs too. Training frontier AI models requires vast computing power. Nvidia’s H100 and Blackwell GPUs can cost between $25,000 and $70,000 per chip depending on configuration and networking. Last week, SpaceX AI announced it would provide Anthropic with access to its Colossus 1 supercomputer that features over 220,000 H100, H200, and next-gen Blackwell GPUs. AI labs also face soaring electricity and networking expenses.

That means revenue growth does not automatically equal profitability. OpenAI reportedly generates billions in revenue, but still aggressively burns cash on compute expenses. Anthropic likely faces similar economics.

The Bigger Story Investors Should Watch

The real takeaway here may not be Anthropic itself. Instead, this could mark the beginning of a new economic cycle where AI platforms scale at speeds traditional software companies never could. Salesforce pioneered cloud software. Anthropic is helping pioneer intelligence-as-a-service. There’s a difference.

Cloud software helped employees work faster. AI increasingly performs the work itself — writing code, summarizing research, automating customer service, and analyzing data. That expands the addressable market dramatically. That said, competition remains fierce.

OpenAI, Google, Meta Platforms, and a growing field of Chinese AI developers are all racing to capture enterprise customers. Pricing pressure could intensify as models become cheaper to run.

Still, sharp investors should recognize what the numbers already tell us: AI adoption is happening faster than prior technology revolutions. The internet took years to monetize at scale. Cloud computing took more than a decade. Generative AI appears to be compressing that cycle into months.

Key Takeaway

In short, Anthropic’s leap from an $87 million run rate to roughly $40 billion in three years may go down as one of the fastest scaling events in business history.

Whether the company ultimately justifies its valuation remains an open question. Profitability, competition, and infrastructure costs all matter. But when all is said and done, the bigger story is impossible to ignore — AI is creating revenue growth curves that make even the SaaS boom look slow.

For investors, that doesn’t mean blindly chasing every AI stock at any price. It does mean understanding that the companies enabling this buildout —  from Nvidia to cloud infrastructure providers and power suppliers — may still be in the early innings of a much larger transformation.