The rapid development and penetration of large models have led to the widespread application of generative artificial intelligence. Various generative artificial intelligence plugins have interacted with the production and life of human society in just a few months, triggering a new wave of intelligence. Just as decades ago, the computer technology and internet revolution led to historic changes in technology and production methods, generative artificial intelligence began to profoundly affect every aspect of people's lives. When we use generative artificial intelligence to efficiently search, generate text, and improve work efficiency, when we feel the freshness of training chat robots, and when we use image generation and audio video generation on top of text, the potential risks of generative artificial intelligence gradually penetrate unknowingly. The duality of technological innovation has always been widely concerned by society, from the data privacy of the Internet to the moral ethics of intelligent robots, and then to the handling of personal privacy and potential fraud risks caused by generative artificial intelligence. How to standardize the application is worth our deep thinking. Users will provide non-public information about individuals, companies, or social organizations to generative artificial intelligence tools. Once the model obtains this data, it may leak information in subsequent generation tasks. In the past, there were concerns about mobile apps and other software observing and calling users' privacy when they were not aware, and in generative artificial intelligence application scenarios, there were also risks of privacy and information leakage. For example, generative artificial intelligence can be used to create information with rich backgrounds and seemingly reliable sources, thereby inducing users to click on malicious links or download malicious software; It can create a rich and three-dimensional social media image, even imitate someone's voice or appearance, guiding rumors and fraud; It can be trained to recommend misleading content to specific users, or in some cases reinforce their existing biases, and so on. Generative artificial intelligence can handle massive amounts of data, but it also brings new challenges to intellectual property protection, personal privacy protection, and combating false information. Actively promoting artificial intelligence governance, the rule of law, and social education are the two key measures to promote innovation and integrity. On the one hand, creating a safe and orderly legal environment and standardizing generative artificial intelligence services. To reduce the spread of false information, it is recommended to explicitly prohibit or restrict its use in specific fields. For example, mass virtual image creation and dialogue; To reduce the risk of privacy leakage, it is recommended to stipulate that generative artificial intelligence requires user consent when using personal data, ensuring that data privacy is protected. Users have the right to decide which data can be used and how the generated content is used; To reduce the destruction of social order and good customs, it is recommended to strictly restrict the creation of hate speech and discriminatory content. It is possible to consider establishing a specialized agency responsible for technical review and certification, evaluating the security, credibility, ethical issues, etc. of generative artificial intelligence systems. At the same time, for the legal liability and compensation mechanism when problems arise in generative artificial intelligence systems, system design can be considered to discuss the allocation of responsibilities among developers, operators, and users. Essentially, any behavior generated by a machine requires the responsibility of an entity, and whether and how generative artificial intelligence can define the responsible entity is an important and urgent issue. Technological Development and Commercial Application of Autonomous Driving