"AI employees" who can automatically write code, gradually bidding farewell to less user-friendly intelligent customer service, and industrial AI controllers that do not require time-consuming and laborious development... With the deepening of artificial intelligence big model capabilities in multiple industries, vertical large model applications in IT, industrial production, finance, service marketing and other industries have entered a period of explosive growth. As a code assistant, Tongyi Lingcode can write code, lead sales, and communicate with people. Nowadays, it is no longer just a skill for showing off muscles. Recently, Alibaba Cloud revealed that it is fully promoting AI programming internally, using Tongyi Lingcode to assist programmers in writing, reading, troubleshooting, optimizing code, etc. The company has also assigned a formal employee ID to Tongyi Lingcode - AI001. "In the future, 20% of the company's code will be written by Tongyi Lingcode, but programmers will still be the core of research and development, and they will have more time to focus on system design and core business development work," said a person related to Alibaba Cloud. Under the traditional development model, programmers need to spend a lot of energy writing repetitive code every day, and carry out basic work such as debugging, optimization, and writing code comments, which greatly squeezes the time for writing core business code. Within Alibaba Cloud, Tongyi Lingcode has taken on the role of code assistant in various development stages. Taking API (Application Programming Interface) development and testing work as an example, Tongyi Lingcode can shorten the manual testing time from tens of minutes to seconds, saving programmers more than 70% of the testing code workload. AI programming tools are breaking the boundaries of programming languages, allowing programmers to easily write code for various programming languages. At present, Tongyi Lingcode, jointly developed by Alibaba Cloud and Tongyi Laboratory, is proficient in over 200 programming languages such as Java, Python, JavaScript, etc. Its generated works have millions of lines of code adopted by programmers every day. From "showing off muscles" to mass landing of digital human sales, it only costs 15% for real people, but can achieve 85% of product transaction volume; In the past, users who spoke to the so-called "intelligent customer service" and needed to find manual customer service due to poor communication have 90% of their problems solved through intelligent customer service self-service; Given a word description, AI can directly generate posters, scripts, videos, and even secondary processing... Recently, Baidu AI Cloud held a large model application product conference, and released a number of large model applications after integrating large model capabilities into the scene. The technology of large models has developed rapidly in the past year, and people's fear and concerns about large models have gradually faded, highlighting the need for rationality and pragmatism. While various large models emerge endlessly, how to make good use of them, bring business growth, achieve cost reduction and efficiency increase, has become a real concern for enterprises. For example, for a long time, intelligent customer service has been repeatedly complained about by consumers. Traditional AI customer service, which cannot understand people's words and has no feelings, has limited help to improve user experience and service efficiency. If the user asks "Why does my car always make a squeaking sound", traditional AI customer service simply cannot understand the colloquial expression of "squeaking". And the intelligent customer service with the ability to access large models will continue to ask users the duration and frequency of the question,