The King is Dead, Long Live the New King
The throne never stays empty, a new king is always found.
Old habits play an important role in human life. Our old knowledge can make us ignore new developments. Similar situations happen with artificial intelligence too. For example, recent numbers show that OpenAI, which was once considered the clear leader in AI, has lost its dominance in the business AI market. Anthropic is quickly rising to take its place. This change is not just about market share. It also shows the deep differences between AI development philosophies and business models.
Anthropic's Birth: A Mission-Focused Split
Anthropic's story begins when a group of researchers left OpenAI. This group included founder CEO Dario Amodei and his sister Daniela Amodei. Dario Amodei played an important role in developing GPT-2 and GPT-3 projects at OpenAI. He doesn't give a clear reason for leaving, but changes in Sam Altman and the company's mission certainly had an effect. It's interesting that the departure happened during the time when OpenAI changed from a non-profit company to a normal company.
Is Open Source a "Wrong Target"?
Anthropic CEO Dario Amodei thinks the "open source" concept is usually misunderstood in the AI sector. He calls it a "wrong target." According to him, there is no situation like "seeing the source code" in artificial intelligence. What matters is whether the model is good and whether it performs better than us in tasks. It doesn't matter if it's open source or not. Running these models is still expensive and complex. This approach shows that Anthropic's business model focuses on model quality and creating value, not on open source competition.
The New King: Anthropic
The "Foundational Model Landscape and Economics" report by Menlo shows Anthropic's success clearly. Anthropic's Claude model has left OpenAI far behind. OpenAI controlled 50% of the business large language model (LLM) market at the end of 2023. Today, it only has 25%. Claude has 42% of the code generation market, which is more than twice OpenAI's 21% share. This trend shows no signs of slowing down.
Anthropic CEO Dario Amodei explains this situation by saying they focus on "business use cases." According to him, business use of AI has much bigger potential compared to OpenAI's consumer-focused approach (ChatGPT) and Google's strategy of integrating into all products. Amodei says that improving a model's ability in biochemistry "from undergraduate level to PhD level" might not mean much for ordinary consumers. But it could be a huge breakthrough for a company like Pfizer.
What About Revenue?
Anthropic's financial performance is also very impressive. Their yearly revenue went from zero to 100 million dollars in 2023, from 100 million to 1 billion dollars in 2024, and from 1 billion to 4.5 billion dollars in the first half of 2025. This proves real "exponential" growth. Anthropic has received almost 20 billion dollars in investment: 3 billion from Google, 8 billion from Amazon, and 3.5 billion from a new round led by Lightseed. They are also Amazon's closest partner in artificial intelligence.
Business Model
"Talent density" is at the center of Amodei's success strategy. Amodei says that Anthropic achieves what other companies do with billions of dollars at much lower cost because they work more efficiently with more talented people. Even when Mark Zuckerberg offered big data centers and talent transfer deals, Anthropic employees refused these offers because they "believe in the mission." The company doesn't compromise on "fair salary principles." According to Amodei, "they are trying to buy something that can't be bought: belief in the mission."
Amodei argues that the perception about his company's profitability is also wrong. He explains that the company appears to be losing money because of the big investments made to train the next model continuously. Each model is profitable on its own, but the billions of dollars spent each year to train the next model makes the company look unprofitable. This situation seems like an unavoidable cost in a constantly developing sector. According to Amodei, this is an investment-return cycle where each model is evaluated like a separate "venture."
Conclusion: The Fast Horse
Anthropic's rise in the business market breaks old patterns. This story reminds us of something: In the technology world, leadership is not won just by being first, but by acting with the right vision at the right time. While OpenAI was running fast, Anthropic chose to run in the right direction. Dario Amodei's words "belief in mission cannot be bought" is a good answer to Silicon Valley's "money solves everything" mentality. But be careful: This story is not finished yet. In the AI race, today's leader can become tomorrow's nostalgia. After all, sometimes the rabbit wins, sometimes the turtle wins.
So which models do you use? Do you also think Anthropic has passed OpenAI? I'm waiting for your comments.
See you in the next article.