AI is one of the most mentioned keywords in 2023, but 6 years ago, China proposed a big strategy. In the 13th National Five-Year Plan in 2016, the country launched the “New Generation AI Development Plan”, which began to be implemented and promoted from 2017. In it, China set the goal of achieving “the world leading degree” by 2025 and become “the world’s major AI innovation center” by 2030, prioritizing making AI a key driver of industrial upgrading and economic transformation.
Illustration of China’s AI-integrated humanoid robot. Photo: CryptoSlate
To realize the goal, China promotes data sharing, encourages cooperation between academia and industry, and attracts talent and foreign investment. The country’s government also encourages AI integration in several fields such as healthcare, transportation and finance. Abundant data sources play an important role in AI development. Initiatives such as the National Engineering Laboratory for Big Data Analytics and Applications have facilitated the use of data in AI research.
Besides promoting development from the government and businesses, a driving force for China’s AI industry comes from the people themselves. In Stanford’s 2022 survey, 78% of Chinese respondents agreed with the statement that AI products and services have more benefits than disadvantages. Meanwhile, only 35% of Americans sampled were optimistic about AI applications.
Great achievement
Statistics announced at the 2020 World Internet Conference at the end of November 2020 in Zhejiang province show that in 2019, China had more than 30,000 patents on AI, an increase of 52.4% compared to the previous year and surpassed the US in the number of new registrations. This milestone marks the first time since 1978 that the US lost its leading position in the number of patents and new inventions.
Before the world focused on AI as it is now, China pioneered the application of technology in real life. In 2017 – the year of the policy to promote artificial intelligence, China installed 20 million AI street surveillance cameras. They are deployed in all areas, from helping to enhance security and support crime prevention to optimizing traffic flow and reducing congestion, optimizing shopping, and developing education.
Typical products include Baidu’s DuerOS integrated into electronic devices such as smart speakers; Tencent’s WeChat AI supports automated chat, content recommendations, and mobile payment services; SenseTime, Yitu Technology develops solutions for visual recognition technology in smart security monitoring systems, facial recognition and medical applications; iFlyTek with natural language voice recognition systems.
New generation AI technology is also becoming popular in self-driving cars. Many cities in China are implementing or testing robotaxi at various stages. For example, in some areas in Beijing, Wuhan or Chongqing, people can call self-driving cars with just a few taps on their smartphones. Baidu and Pony.ai have been licensed to operate robotaxi at competitive prices compared to traditional services.
Simulation of a digital image of a woman thanks to AI technology. Photo: EPA-EFE
This year, during the generative AI fever, companies in this country such as Baidu, Alibaba and ByteDance also continuously launched large language models (LLM) to compete with systems from the West.
In March, Baidu launched chatbot Ernie . A month later, it was Alibaba’s turn to announce Tongyi Qianwen , which is likened to “the Chinese version of ChatGPT”. SenseTime, backed by SoftBank, also introduced its large language model SenseNova and integrated it into a chatbot called SenseChat.
In early November, Beijing-based AI startup Baichuan announced that its self-developed AI model Baichuan2-192k was able to process about 350,000 Chinese characters and became the world’s most powerful model in the world. processing long text commands.
Also in November, 01.AI, a startup founded by computer scientist Lee Kai-Fu in March, became a technology unicorn, or was valued at more than a billion USD after announcing an LLM with named Yi-34B. The platform, which is available in English and Chinese, is said to outperform leading open source models currently on the market, including Meta’s Llama 2, thanks to its ability to handle 100 billion parameters, the largest of its kind. with current open source models and can compete directly with OpenAI’s GPT-4, according to Straitstimes .
“Our proprietary model will be benchmarked with GPT-4,” said Kai-Fu.
Reuters quoted Baidu CEO Robin Li as saying that, as of early September, more than 70 major language models with more than a billion parameters had been released in China. According to brokerage company CLSA, by the end of October, China had at least 130 LLMs, accounting for 40% of the global total and only behind the US with 50% market share. In addition, companies also publish dozens of LLMs specific to certain fields.
Besides the AI model, China is also starting to become self-sufficient in hardware. In mid-November, Tsinghua University announced that it had successfully created a new AI chip that is 3.7 times more powerful than Nvidia’s A100 in computer vision tasks. Called ACCEL, the chip leverages photonic and sequential computing capabilities in a specialized architecture that relies on light and uses photons to handle high-speed information transfer.
According to a veteran figure in the chip field told SCMP , Tsinghua University’s achievements will motivate China to research optical computing chips to replace current electronic chips, as well as new AI chips in the future. . “In some cases, performance from photonic computing is much stronger than from electronic computing,” the person said.
“This year, China’s AI industry witnessed unprecedented progress. Thanks to policy support, China entered a period of rapid development. China’s major AI models are competing fairly with Western models,” Global Times quoted independent technology expert Liu Dingding.
Bao Lam
EN 
VI
Leave a Reply
You must be logged in to post a comment.