Anthropic is not just an opponent of OpenAI, it's more like a decade late counterattack! If you only see Anthropic as a competitor to OpenAI, then you are underestimating this matter. It's more like a group of people who came out of OpenAI years ago, holding their breath and finally coming back to settle accounts. This is not an ordinary entrepreneurial story. This is an old man turning over. It's a long-standing grudge. It was also a counterattack that dragged on for many years. 1. Anthropic is not 'wanting to be another AI company', but a group of people who are really fed up with OpenAI. Outside, Anthropic is often portrayed as a safer, more restrained, and more idealistic alternative. That's right. But that's not all. The more direct reason is that Dario Amodei's group of people could no longer tolerate OpenAI's internal gameplay. The earliest contradictions still seem like ideological differences. For example, how should cutting-edge AI information be made public. Should we first inform the public or communicate with the government. But as we move forward, it becomes less and less of a problem. Because what truly turns people upside down is never just opinions. But you find that the people you work with every day talk about their mission while playing with something else. 2. What really drives people away is not the technical route, but OpenAI's system of power and credit allocation. Dario is not an outsider at OpenAI. On the contrary, he was one of the most central figures of the GPT-2 and GPT-3 eras. That's the problem. He did the job. He has a share in the results. But as he progressed, he felt increasingly marginalized in the most critical positions. Who speaks on behalf of the company. Who goes out to meet big shots. Who holds the power of internal discourse. Who makes the decision at critical moments. These things seem like small things, In fact, in top companies, everything is the most deadly thing. What's even more disgusting is that people like Altman are best at soothing both sides and leaving gaps on both sides. While telling Dario, everyone has no confidence in Brockman and Ilya. I promised the other side that I still have the right to move Dario. This is not management anymore. This is a mine burial. No one is completely offended. Everyone should stabilize first. In the end, all contradictions exploded together. 3. What makes OpenAI most popular is not the power struggle itself, but its ability to pretend. This may be the most unbearable part for the Anthropic group. OpenAI's greatest selling point is never just models. But rather posture. Safe. Responsibility. mission. Public interest. The future of humanity. One set is told beautifully. But what is inside? Still grabbing power. Still excluded. Still throwing the blame at each other. Still competing for control and credit. This is the most infuriating thing. Because if you were to say it nakedly from the beginning: Laozi is a commercial company that seeks power and victory. On the contrary, it is not so repulsive. The most annoying thing is that my mouth is full of idealism, Behind it all is office politics. That's also why later Dario criticized OpenAI internally within Anthropic, which is no longer the kind of criticism used by ordinary competitors. What tobacco company. What hypocrisy. What kind of pattern lies. This is no longer PR language. This is a true emotion that has been suppressed for many years and cannot be held back. 4. Anthropic's later strategy of 'we are the healthier alternative' was essentially aimed at OpenAI. Do you think that's just brand positioning? No. That was a positioning made with resentment. What has Anthropic been trying to package itself into over the years? More stable. More secure. More trustworthy. It's not like OpenAI. This is not just for the market. Also for drawing lines. The meaning is very clear: Aren't you the most responsible, safe, and restrained? Then let's be the ones who really turn these things into the corporate atmosphere. You are responsible for the performance. We are responsible for taking away this flag. So many actions in Anthropic seem to be doing themselves on the surface. In fact, it is also working against OpenAI. OpenAI is more like the kind of company that wants to monopolize, eat out the commercial dividends, and continue to occupy the moral high ground, The more space Anthropic has to present itself as the 'more normal' choice. 5. So the most critical dark line in the AI industry today is not any technical route, but rather the fact that these people never reconciled back then. Many people always want to make the relationship between OpenAI and Anthropic very rational. What is Security vs. Product. What is closure vs. restraint. What is the speed of commercialization versus public interest. All of these are correct. But if you take away the emotions, the story becomes tasteless. The more authentic version is: The success of these two companies today is not only due to different routes, But a group of people who used to work together, but later completely lost faith in each other. Not believing in ability. Do not believe in motives. I don't believe in character. So Anthropic didn't just emerge from the ground. It is the company form that was created by the long-term internal cracks of OpenAI and eventually exploded.