DeepSeek said debut training cost about $6 million vs $100m for GPT-4
Anthropic and OpenAI have claimed that some Chinese AI firms, including DeepSeek, essentially copied their AI models, according to BBC News. Both US firms are poised to list on the public stock market in the coming months, BBC News reported.
The copying claims sit against a backdrop of rapid adoption of DeepSeek’s technology. DeepSeek’s models are open source — free to access and less expensive to use — and the firm has grown in popularity since its initial public release in 2025, with demand at times so strong that the company’s servers have gone down, BBC News reported. DeepSeek’s free chatbot app ranks among the most-downloaded apps in the Apple App Store, and US technology companies — including Snowflake, Vercel and OpenRouter — incorporate DeepSeek models into their offerings to customers, the BBC reported. Startups have flocked to open-source models like DeepSeek.
The cost gap between DeepSeek and major US AI firms is substantial. For developers, using DeepSeek costs a fraction of what a closed model such as Google’s Gemini, Anthropic’s Claude, or OpenAI’s Codex would, BBC News reported. Last year, DeepSeek said it took about $6 million (£4.5 million) to train the debut version of its model, while the costs associated with OpenAI’s GPT-4 were roughly $100 million. Those disclosures sent tech stocks tumbling, according to BBC News.
DeepSeek released its most recent version last week. Users of the model have praised it online, saying it beats even the most advanced AI tools from OpenAI on common tasks such as computer software coding, BBC News reported.
The Department of Homeland Security on Monday, Oct. 19, made public an order banning DeepSeek at civilian federal agencies, and any technology built with the aid of its AI models. That ban took effect at the end of July, according to BBC News. The order covers the Department of the Treasury, the Department of State and the Department of Justice; agencies within the Department of Defense are excluded. The DHS notice did not detail the specific security concerns behind the ban. DHS, DeepSeek, and the Chinese embassy in Washington, D.C., have been contacted for comment by BBC News.
The ban follows criticism of Chinese AI from the Trump administration and from American firms. White House tech lead Michael Kratsios wrote earlier this year in an internal memo that “malicious exploitation” of American AI firms had been occurring, mainly by rival firms in China, BBC News reported. President Donald Trump has made clear that he sees China as the key US rival in the development of artificial intelligence, a technology he has described as crucial to the future of the country, according to BBC News.