According to Inc, Anthropic’s annual run-rate revenue has exploded from $87 million to a staggering $7 billion as of October 2025, with projections to hit $9 billion by year’s end. This growth is largely fueled by entrepreneurs like Wyndly CEO Aakash Shah, who now uses Claude to complete investor updates in an afternoon instead of days. The company, founded in May 2021 by ex-OpenAI research head Dario Amodei and six colleagues, introduced Claude in March 2023. Its ability to write code has spawned a movement called “vibe coding,” where users describe features in plain language and watch the AI build them. This has made Claude an integral partner for AI coding companies like Replit and Cursor, effectively positioning it as the technical co-founder for countless startups.
The OpenAI Exodus That Built a Rival
Here’s the thing about origin stories: they often get polished into myth. But the Anthropic one is pretty stark. Dario Amodei and his team weren’t just some researchers leaving a big lab. They were core architects of GPT-3 at OpenAI. That’s like the lead engineers for the iPhone leaving Apple right after its launch to start their own phone company because they thought Apple would be reckless with it. Amodei’s stated reason—that OpenAI wasn’t “particularly thoughtful” about responsible scaling—is a massive, career-betting critique. It frames Anthropic not just as another AI vendor, but as the “responsible” alternative. That’s a powerful brand, especially when selling to enterprises who are terrified of AI going off the rails. But you have to wonder: is the safety-first stance a genuine differentiator, or just a brilliant marketing angle in a market dominated by OpenAI’s consumer-facing ChatGPT?
The Vibe Coding Gold Rush
So “vibe coding” is a thing now, even if Amodei thinks the term is “woo-woo.” And look, the anecdote about Aakash Shah is compelling. Going from a hobby coder struggling for hours to fixing problems in minutes? That’s a powerful drug. It democratizes a skill that has been a major bottleneck for decades. But I’m skeptical. This feels like the early days of website builders promising “no code needed!” They work wonders for specific, standard tasks, but the moment you need something truly custom or complex, you hit a wall. Will startups built on vibe-coded foundations be able to scale? Or will they end up with a tangled mess of AI-generated code that no human fully understands? The feedback loop theory from Anthropic’s Alex Albert makes sense—code is testable. But debugging an AI’s spaghetti code when you yourself don’t know programming seems like a fresh kind of hell.
The Enterprise Bet and the Revenue Machine
This is where Anthropic’s strategy gets really smart. While OpenAI captured the world’s imagination with ChatGPT, Anthropic went straight for the enterprise wallet. They built a tool that solves a very expensive business problem: the shortage and cost of skilled developers. Turning every founder and product manager into a pseudo-developer is a trillion-dollar proposition. The revenue numbers—$87M to $7B—are almost absurd. They indicate Anthropic isn’t just selling API calls; they’re embedding themselves into the core operations of other tech companies. But that kind of hyper-growth brings its own dangers. Can their infrastructure and support scale as fast as their sales? And what happens when the next, better coding model comes along from a competitor? Their whole empire is built on being the best at this one, specific, incredibly valuable task. It’s a fantastic position to be in, but it’s also a very narrow tightrope to walk.
Is Claude Really a Co-Founder?
Calling Claude a “co-founder” is a great headline, but it’s also a bit of a fantasy. A co-founder has judgment, context, and skin in the game. An AI has none of those. It’s an incredibly powerful tool, arguably the most powerful tool for software creation ever invented. But a tool nonetheless. The real story isn’t about AI replacing founders; it’s about AI dramatically lowering the activation energy to build and test software ideas. The risk is that this leads to a tsunami of half-baked, poorly architected apps. The opportunity is that it lets innovators focus on the *what* and the *why* instead of getting bogged down in the *how*. Anthropic’s bet is that the latter will win out, and so far, the market is screaming that they’re right. But let’s not confuse a brilliant assistant with a true partner. The vision still has to come from a human. At least for now.
