As the discourse around AI-based chatbots has heated up after the public launch of OpenAI’s ChatGPT, we look at China’s reaction both from the angle of businesses and regulators.
The debate around chatbots has once again resurfaced since OpenAI launched ChatGPT, a new artificial intelligence (AI) tool capable of writing essays and code. It took ChatGPT just five days to reach one million users. The app has also become the fastest-growing consumer app in history since its release in November 2022.
With an increasing number of businesses using chatbots in their day-to-day practices to automate many areas of the customer experience, any upgrades to the capability of these systems based on AI have quickly become hot topics.
Despite concerns about ChatGPT’s capabilities, the implications for future business opportunities have gotten all major tech companies in China scrambling to create their own versions of the AI-powered chatbot that has the U.S. tech industry buzzing – case in point, Google’s ‘Bard,’ a ChatGPT-like chatbot which will supposedly expand the engine’s own search tools, but is already displaying major flaws.
In this article, we analyze the ChatGPT phenomenon, China’s response to the technology, and its significance for the future of work.
What is ChatGPT?
ChatGPT is a chatbot based on a language model developed by U.S. firm OpenAI and backed by Microsoft Corp that has been trained on a massive amount of text data from the internet. The goal of ChatGPT is to generate human-like responses to natural language questions and prompts, allowing it to carry out conversations with users in a coherent and natural manner.
The chatbot is based on the transformer architecture and utilizes the GPT-3 (Generative Pretrained Transformer 3) algorithm. A transformer is a deep learning model that adopts the self-attention mechanism and weighs the importance of each component of the input data differently.
Accordingly, ChatGPT has been trained on a diverse range of text data, including books, news articles, websites, and more, which allows it to have a broad understanding of a wide range of topics. In other words, this model can understand the context and generate contextually appropriate responses.
ChatGPT’s applications and uses
In terms of applications, ChatGPT can be utilized in various industries, such as customer service, education, and entertainment. One of its primary applications is in the field of Natural Language Processing (NLP). The model can be used to generate text based on given inputs, making it ideal for tasks, such as language translation, text summarization, and answering questions. It has also been used to build conversational AI systems, such as chatbots, which can be used in customer service and support applications.
Another application of ChatGPT is in the field of content creation. The model can be used to generate articles, stories, poems, and other forms of written content. This has been especially useful for organizations and individuals that need to create a large amount of content in a short period of time.
The chatbot can also be used for data analysis and research, as it can be trained on a large corpus of data to extract insights and trends. In addition, the model can be used in the field of education, for example, to generate personalized study materials for students.
All in all, the versatility of ChatGPT makes it a valuable tool for a wide range of industries and applications.
How is China viewing ChatGPT?
ChatGPT is generating a lot of curiosity in China, as local businesses are purportedly hurrying to incorporate chatbot technology into their products and services. The model is competent in the use of the Chinese language, which facilitates its adoption throughout the country.
However, Chinese users are currently unable to access ChatGPT. Although neither OpenAI nor ChatGPT are prohibited by Chinese regulations, OpenAI does not allow users based in mainland China, Hong Kong, Iran, Russia, and some regions of Africa to register on the platform.
Still, young Chinese users have discovered workarounds, such as using VPNs or foreign phone numbers to access the OpenAI website and test ChatGPT in creative ways. Some users have used the chatbot as a movie reviewer, job counselor, financial manager, or dream translator to then share their feedback online about their experience.
ChatGPT’s limitations and critiques
Other than the ethical nature of the discourse surrounding the use of AI-based language models, ChatGPT is not exempt from technical limitations either. First, as already pointed out by several experts in the field, ChatGPT is unable to fully comprehend the nuanced nature of spoken and written human language. It is just taught to produce words depending on input, but it cannot fully understand the meaning of those words. This indicates that any answers it elicits are probably going to be devoid of creativity and true understanding.
Another major issue regards fact-checking: several experiments revealed that ChatGPT repeatedly provides false information and invents non-existing sources and links to back it up. Thus, every content created using ChatGPT should always be submitted for rigorous fact-checking.
How will ChatGPT change the way we work?
Generative AI, or artificial intelligence that can generate original text, audio, and visual content in response to user input, is already raising concerns that it may replace a variety of jobs. The most widespread idea, however, is to use such tools as ‘assistants’ and make certain jobs more accessible to everybody. BuzzFeed, for example, will reportedly employ the chatbot created by OpenAI to improve its quizzes and personalize some material, which would then be revised by its internal team.
After all, we have already seen how other forms of AI have been applied to different settings. ITV has produced a comedy show based on deep fake depictions of celebrities, and Hollywood is using AI to de-age performers.
A crucial aspect of adaptation is determining how to restructure our economic systems so that the entire working population can be effectively utilized. This entails implementing systemic solutions that don’t merely redistribute tasks between humans and machines.
Who are the major competitors to ChatGPT in China?
Baidu’s shares rose to an eleven-month high after the company announced its plans to introduce the ChatGPT-inspired system called ‘Ernie Bot’, based on technology that has been developing since 2019. The company plans to finish internal testing in March before making the chatbot available to the public.
In response to Baidu’s announcement, the tech giant Alibaba stated that it is internally testing a ChatGPT-style tool. However, the company did not provide information on the launch date or the name of the application. Tencent has joined the race as well, confirming its plans for ChatGPT-style and AI-generated content and stating that relevant research is ongoing.
Meanwhile, online retailer JD.com plans to integrate natural language processing technologies like ChatGPT into its services, and gaming company NetEase is researching the integration of AI-generated content into its education division.
Finally, reports from Chinese media claim that Byte Dance’s AI lab has launched research initiatives to support its virtual reality branch, ‘Pico’.
AI-related stocks listed in mainland China have risen considerably in the first two months of 2023, as a direct result of the growing public interest in ChatGPT. The value of the Shanghai-listed shares of Beijing Haitian Ruisheng Science has increased by nearly 205 percent so far this year. Similarly, Hanwang Technology has grown by 124 percent and CloudWalk Technology Company by 105 percent.
How do Chinese regulators view ChatGPT and AI-based applications?
Chinese regulators, who introduced some guidelines last year to tighten oversight of deepfake technology, have not yet commented on ChatGPT. Nevertheless, China’s state media has warned investors about stock market dangers amid a craze around local ChatGPT-concept stocks.
The reaction of Chinese regulators to the chatbot’s gold rush is fact-based., for example, lost almost 10 percent of its stock shares at the beginning of February 2023 after releasing its own AI-powered chatbot, ‘Bard‘, which was presented as a valuable competitor to Microsoft-backed ChatGPT. In particular, Google’s stock crashed after Bard gave the wrong answer to a question. Supposedly, the main difference between Bard and ChatGPT is that the former should be able to navigate the internet in real-time to provide an updated answer in a human-like language form.
Bard’s factual error in its first demo is a sign that investors should act cautiously when it comes to early developments of AI chatbots.
Indeed, China has taken the lead over AI regulations. Several large-scale initiatives have been put in place to both encourage and oversee the growth of the AI sector, including Made in China 2025, the Action Outline for Promoting the Development of Big Data (2015), and the Next Generation Artificial Intelligence Development Plan (2017). In recent years, the country has also accelerated the pace of enacting explicit laws governing the ethics of AI businesses and their algorithms.
Meanwhile, Beijing has also been actively promoting the growth of its data economy, defined as a ‘factor of production’, laying out a plan for constructing the infrastructure necessary to support the development of basic data systems.
It is possible that the Chinese government may regulate AI-based language models like ChatGPT as part of its broader efforts to regulate the technology industry. Companies like OpenAI may need to comply with any relevant regulations in order to operate in China. However, the specific nature and extent of these regulations are currently unclear.
What are the challenges for China businesses to adopt ChatGPT-type technology?
In their application of domestic ChatGPT-like chatbots, Chinese businesses face a dilemma. On the one hand, they need to persuade customers and investors that they are not falling behind in the development of the new technology. On the other hand, they also need to avoid being perceived by regulatory bodies as creating new products, services, and business models that may create security concerns.
The same is true for foreign investors and businesses interested in opportunities in this sector when approaching China’s unique situation, especially concerning its cyberspace.
What is quite evident is that, given China’s distinct internet environment, such a balancing act will result in the use of ChatGPT-style technology that differs from what we are witnessing in the rest of the world, particularly western countries like the U.S.
Investing in China’s AI sector
The Chinese AI industry is growing rapidly and is considered a key area for development by the government. Despite some restrictions, the country is motivated to become more self-sufficient in AI technology and is still open to foreign investment and talent. Industry experts at IDC predict that China’s AI investment will reach US$26.69 billion in 2026, accounting for about 8.9 percent of the total global investment, and ranking second worldwide.
Companies need to be flexible and adjust to the changing business and political climate, or they may face a decline. By taking advantage of China’s supportive policies, companies can invest more in R&D and talent. There are many opportunities for business development and scalable enterprises, as well as incentives for R&D innovation.
China Briefing is written and produced by Dezan Shira & Associates. The practice assists foreign investors into China and has done so since 1992 through offices in Beijing, Tianjin, Dalian, Qingdao, Shanghai, Hangzhou, Ningbo, Suzhou, Guangzhou, Dongguan, Zhongshan, Shenzhen, and Hong Kong. Please contact the firm for assistance in China at email@example.com.
Dezan Shira & Associates has offices in Vietnam, Indonesia, Singapore, United States, Germany, Italy, India, and Russia, in addition to our trade research facilities along the Belt & Road Initiative. We also have partner firms assisting foreign investors in The Philippines, Malaysia, Thailand, Bangladesh.