The Zhitong Finance App learned that on February 26, the first Chinese original Wensheng video AI animation series “Ode to a Thousand Autumn Poems” was broadcast on the CCTV-1 comprehensive channel. Industry insiders analyzed that with the launch of the first AI animation in China, the implementation of Wensheng's video application is expected to accelerate, reshape the production process in the film and television industry, and further promote cost reduction and efficiency in film and television animation production, game development, and advertising design.

According to the Shanghai AI Lab, “Ode to a Thousand Autumn Poems” is the first AI animation broadcast on TV in China. Using big model technology, from art design to motion effect generation, to post-production, the film is based on AIGC (Generative Artificial Intelligence) assisted production of the “CCTV Audio Media Big Model”. It highly recreates the characters, scenes, and props in ancient Chinese poetry. It is a “two-way journey” of Chinese poetry and artificial intelligence.

The director of “Ode to a Thousand Autumn Poems” said that AI is not intended to replace humans, but rather helps humans improve efficiency. For example, with traditional pure manual production, our team can only do one episode a month; now we can do three episodes in a month. According to reports, under the same budget, according to the traditional animation production process, “Ode to a Thousand Autumn Poems” will take at least 8 months. With the power of AIGC, the production cycle was shortened to 4 months.

Shenyang, a professor and doctoral supervisor at the School of Journalism at Tsinghua University, said that with the rapid iteration of AI-generated video technology, AI video has entered a period of major explosion. This year, we'll probably be able to see longer movies that don't require actors.

As an important branch of generative AI, Wensheng Video Technology will make a big difference in the media, film and television industry. According to information, not only will film and television giants such as Pixar and Disney Animation Studios participate in the GTC AI event that Nvidia will host next month (March 18 to March 21), but the application of generative AI in film and television will also be highlighted. For example, the exhibition hall will display a giant multi-sensory interactive device by world-renowned AI artist Refikanadol, as well as a demonstration of Cuebric, a generative AI tool designed by Seyhanlee for filmmakers.

Guotai Junan believes that the AI video generation model is iterating faster than expected, and the value of the animation industry is expected to be reshaped. The video content industry chain such as animation can be divided into three major links: upstream IP creativity, midstream filming and editing, and downstream content distribution. The AI video generation model may have the greatest impact on the midstream link, which accounts for the highest proportion of repetitive labor. As the value of the midstream chain is gradually replaced by AI, the importance of upstream IP ideas is expected to continue to be highlighted.

China Merchants Securities said that Sora is now a landmark new model in the field of video generation, providing solutions for future AI to be deeply involved in content creation. Domestic efforts are being made simultaneously, and upstream film and television IP, filming, post-production, and quality control may all usher in changes. Shanghai Film, Huace Film and Television, Aofei Entertainment, Jiecheng Co., Ltd., Tom Cat, Reading Group, Chinese Online, etc. with rich IP or video content materials may benefit from the increase in demand for upstream IP and materials brought about by technological progress; Optical Media, Beijing Culture, Zhejiang Film, China Film, etc. that are involved in film and television content production and actively embrace AI technology may also benefit from increased production capacity. Considering that the supply of film and television content is expected to grow in the future, cinema companies such as Wanda Film, Happy Blue Ocean, and Hengdian Studios are also expected to benefit.

Related concept stocks:

Reading Group (00772): On June 19 of last year, Yuewen Group began an organizational structure upgrade to establish four major divisions. One of them is the Intelligence and Platform R&D Division, which is responsible for technical attacks such as AI models and their transformation into AIGC scenario applications. More than a month later, Reading Paper released the vertical industry model “Reading Pen”, and the application product “Writer's Assistant Wonderful Pen Edition” based on this big model.

Alibaba Pictures (01060): The film “The Girl Who Disappeared”, co-produced and promoted by Alibaba Pictures, grossed over 3.5 billion yuan, making it the 2023 Dragon Boat Festival box office winner and 2023 summer runner-up; its co-produced film “All or Nothing” grossed more than 3.8 billion yuan, making it the 2023 summer box office winner.

Zhihu (02390): The “Intelligent Chart AI” model was developed on the basis of the mature big model CPM-Bee. It has 100 billion parameters, and is claimed to have stronger logical reasoning capabilities and faster training and reasoning speed. The big model was collaboratively released by Zhihu and Face Wall Technology at the “Zhihu Discovery Conference” on April 13 last year. At the same time, a closed beta was announced. At the same time, an AI-based “Hot List Summary” function was launched. At the same time, the AI-based “Hot List Summary” function was launched, which can capture, sort, and aggregate answers to questions on the Zhihu Hot List, and present a summary of the answers to users.