The Chinese substitute for American-made AI models, DeepSeek, has swept the industry.
A Chinese company has entered the artificial intelligence chatbot market. It is upending American dominance in AI-based generative chatbot models by competing with industry titans like OpenAI, Gemini, ClaudeAI, and others. The Chinese substitute for American-made AI models, DeepSeek, has swept the industry.
The founders of DeepSeek built the model for just $5.6 million. This is only a small portion of the funds that industry titans like ClaudeAI, Google’s Gemini, and OpenAI utilise to develop their models.
An old video of Sam Altman, the founder of OpenAI, calling it “hopeless” to construct an AI model on a $10 million budget has gone viral. “How is a small, smart team with a budget of $10 million could build something substantial within AI?” was the question posed to Mr. Altman during a 2023 event in India.
In hindsight, this is somewhat comical.
In 2023, Altman was questioned in India about how a small, intelligent team could create something significant in AI on a $10 million budget.
— January 28, 2025, Arnaud Bertrand (@RnaudBertrand)
His response was: “It’s totally hopeless to compete with us on training foundation models” . The American chipmaker Nvidia has lost about $600 billion in value in a single day, the largest loss for a single company in market history, while DeepSeek’s chatbot has emerged as the most popular free software on Apple’s US software Store.
When DeepSeek revealed in a study last month that training its DeepSeek-V3 model with the less powerful Nvidia H800 chips took less than $6 million in processing power, the company garnered popularity.
According to a post on DeepSeek’s official WeChat account, the DeepSeek-R1 model, which was released last week, is 20–50 times more cost-effective to utilise than OpenAI’s o1 model, depending on the workload.
On X, Mr. Altman commended his competitor’s RI model, saying, “DeepSeek’s r1 is an impressive model, particularly around what they’re able to deliver for the price.”
“But mostly we are excited to continue to execute on our research roadmap and believe more computing is more important now than ever before to succeed at our mission,” Altman said.
Given DeepSeek’s widespread appeal and inexpensive manufacturing, the American story of investing billions of dollars to develop generative AI models has come under scrutiny.
