WEBDESK: Chinese AI startup DeepSeek is sharing some of its AI model code to back up its open-source development.
Universities in China have launched AI courses based on the Chinese startups models since the breakthrough, which has been called the “Sputnik moment” for China. The government wants more scientific and technological innovation in schools to boost the economy.
DeepSeek, based in Hangzhou, has got thumbs up from Silicon Valley and US tech engineers with its models DeepSeek-V3 and DeepSeek-R1 compared to OpenAI and Meta’s top AI systems.
Shenzhen University just launched an AI course using the AI startups tech. The course will teach students about AI, security, privacy and ethics to balance innovation with ethics.
The open-source code released by the Chinese startup will support and improve its publicly available AI models. This follows the company’s latest algorithm release, Native Sparse Attention (NSA) which speeds up AI for long text processing.
“Having others follow your innovation gives a great sense of accomplishment,” officials said in July.
“In fact, open source is more of a cultural behaviour than a commercial one, and contributing to it earns us respect” he added.
Since last month, DeepSeek’s chatbot has become China’s most popular AI service with 22.2 million daily active users—beating Douban’s 16.95 million users.
Read more: Elon Musk asks federal workers to justify their work or lose their jobs
