On February 13, U.S. media reported: "OpenAI has warned U.S. lawmakers that its Chinese competitor DeepSeek is using unfair and increasingly sophisticated methods to extract results from the leading American artificial intelligence models to train its next-generation product of the breakthrough chatbot R1."
OpenAI comes knocking: AI can't compete, so they start spreading mud! Bloomberg said OpenAI called on the U.S. Congress, accusing China's DeepSeek of using improper means to learn technology. This seems more like the anxiety of a pioneer rather than a matter of technological justice. Looking back at the history of technological development, from semiconductors to the Internet, Western tech giants have always used the rhetoric of "violating regulations and stealing secrets" to suppress later competitors. Now, DeepSeek has gained an advantage with high efficiency and low cost, and R1 has shown impressive performance, breaking the monopoly with hard power.
Model learning and training with public data are common practices in the global industry. U.S. companies also did the same in the past. Now, turning commercial competition into politics shows that technical advantages are no longer there, and they have to rely on rules instead. True technological competition relies on innovation and efficiency, not on smearing and suppression. Stigmatizing normal technological exchanges will only hinder global AI development. Chinese AI speaks with its own strength, and time will prove everything.
Original article: toutiao.com/article/1857002038343815/
Statement: The article represents the views of the author alone.