ChatGPT developed by OpenAI has attracted lots of attention. ChatGPT, based on deep learning and reinforcement learning with human feedback, is a large generative AI language model, which benefited from pre-training concerning massive quantity of data, can understand users' order and generate natural language. Compared with other AI language model, ChatGPT can carry out endless conservation with users, ask users for more information initially, generate human-like content and remember users' generated content. As a masterpiece of generative AI, ChatGPT may become a lever to leverage the fourth industrial revolution through its powerful natural language processing ability, and may bring profound changes to human society, including:it may promote industrial transformation and upgrading, improve labor productivity, improve production tools, and optimize production decision-making by opening up a field for intelligent information processing to transform production methods; it helps to sort out research status, organize research ideas, process research data, and improve research results to transform research paradigms; it empowers the education subject, the education object, and the education process to transform education forms; it reconstruct labor life, consumption life, and spiritual life to transform lifestyles. The development of science and technology is a two-edged sword. ChatGPT will also bring us a series of risks and hidden dangers. It may breed legal issues such as intellectual property infringement, information theft, and fraud, breed ideological problems such as loosing ideological security defense lines, degrading independent thinking ability, and harming value shaping, and breed social problems such as unemployment of workers and increased probability of extreme events. To take advantage of generative AI represented by ChatGPT and make it benefit state building, national rejuvenation and people's daily life, we should firstly adhere to a new system for mobilizing the resources nationwide to achieve breakthroughs in core technologies in key fields, collaborate with "effective markets" and "promising governments", strengthen talent team construction, and mobilize the resources nationwide to develop generative AI independently; secondly, guided by practice, we should rely on the implementation of the empowerment-application mechanism to attract and support the active application of generative AI by the entire society; thirdly, we should strengthen legislation and law enforcement to regulate the development and application of generative AI; fourthly, we should reinforce the implementation of ideological guidance and comprehensive quality enhancement to help people cope with problems generated by generative AI.