Chinese artificial intelligence models are around six months behind those being used in the United States but could soon outstrip them in terms of adoption, says Kai-Fu Lee, the former president of Google China.
The less-advanced Chinese models are roughly 15 months behind their US counterparts, but leading large language models (LLMs) used by Chinese companies are between six and nine months behind, Lee said at the AVCJ Private Equity Forum in China, according to a Sept. 11. CNBC report.
Lee, the founder of startup 01.AI and venture capital firm Sinovation Ventures, said AI apps in China could soon outstrip their competitors in the US, noting that the cost of training AI models had been significantly reduced in recent months.
China will lead on AI apps
“Apps, I would predict, by early next year will proliferate in China much faster than in the US,” Lee said but explained that it’s still unclear whether these apps will be built by small or large firms.
It could take somewhere between five to eight years to bring consumer-ready generative AI capabilities to the level of a “super app,” he added — a single application that performs a wide range of tasks.
AI technology will need entirely new devices, Lee believed as current smartphones won’t be enough to handle the requirements for AI.
“The right device ought to be always on, always listening.”
Lee’s comments concerning permanently observant AI devices come after the July release of an AI-powered wearable called the “Friend” necklace, a device marketed as being a virtual companion that constantly listens to its wearer.
Much like the US, the AI sector in China has witnessed a mass of interest from leading tech firms.
Mainstay Chinese firms Alibaba and Tencent have released several iterations of their own in-house AI models and applications, as well as funneling billions in funding towards smaller AI startups.
Tencent unveiled its own LLM called “Hunyuan” in September last year, the Chinese tech giant’s answer to OpenAI’s chatbot ChatGPT.
Hunyuan has since been integrated throughout the firm’s company ecosystem, including in its cloud computing, marketing and gaming divisions.
