Renowned computer scientist and venture capitalist, Kai-Fu Lee, has launched his new artificial intelligence startup, 01.AI (Lingyi Wanwu in Chinese), revealing his plans to develop an AI 2.0 platform and applications. The Beijing-based company has chosen to take on the challenge of building its own large language model (LLM), considering it a historical opportunity for China.
In a post shared on the WeChat account of Lee’s VC firm, Sinovation Ventures, the startup explained its aim to create a domestically-grown model capable of producing products similar to OpenAI’s ChatGPT. Lee believes that once China has truly native, high-quality LLMs, the country will see a multitude of high-quality and creative applications, comparable to the era of mobile internet.
01.AI has outlined its model training strategy, consisting of seven major modules, including pre-training, post-training, AI infrastructure, and multi-model technology. The company aims to bring in top-notch technical experts for each module to develop an LLM with enhanced capabilities.
The success of OpenAI’s ChatGPT has sparked a race in China, with tech majors, startups, and research institutions vying to create similar models. According to data from a state-backed scientific institution, China already has at least 79 LLMs with parameters exceeding 1 billion as of late May. This indicates the growing interest and investment in the field within the country.
With the launch of 01.AI, Kai-Fu Lee aims to propel China forward in the realm of artificial intelligence, recognizing the tremendous potential and significance of large language models. By developing its own LLM, China can tap into a historical opportunity and unlock a new wave of innovative applications. The competition in the field is fierce, and only time will tell which company will emerge as the leader in this rapidly evolving sector.
