on may 14, Beijing time, OpenAI, an American artificial intelligence company, released a new flagship model "GPT-4o" in the "spring update" activity. the model supports any combination of text, audio and image input and can generate any combination of text, audio and image output. Compared with existing models, GPT-4o is particularly good at image and audio understanding, and can react to audio input within 232 milliseconds, which is similar to the reaction time of humans in conversation.

The "o" in the name of the GPT-4o stands for Omni, which means "all-round", highlighting its multi-functional characteristics, and based on the GPT-4o ChatGPT, all users can use it for free, while the capacity limit of paid users is five times that of free users.

After the release of the GPT-4o, the expectation and confidence in AI technology in the market will be further improved.

Guosheng securities research paper believes that GPT-4o is a major progress towards more natural human-computer interaction. the new function has brought new multi-modal interaction capability. through the new end-to-end model, a new breakthrough in experience has been realized. it is expected to maximize user experience in various terminals and make good use of intelligent terminal Agent, robots and other directions.

With the continuous development and application of AI technology, the multi-modal AI-related industry chain will usher in more opportunities and growth space. Stimulated by this, the multi-modal AI concept stocks associated with A- shares have also risen.

according to the data, as of the close of trading on may 14, the multimodal AI index (886062) rose 2.61 percent to 827.89 points, with Tom cat (300459.SZ), hongbo medicine (301230.SZ), voice shares (003004.SZ) and other multimodal AI concept stocks rising, Jiangnan chemical (002226.SZ), meiansen (300275.SZ), S300029, etc. zxS.

As the earliest experimental field for AI-related technologies, the game sector once again led the rise. Flush Mobile Game Index (885457) rose 3.65 percent to 1906.67 points, with Tom Cat, Caesar Culture (002425.SZ), Dasheng Culture (600892.SH), Guohua Network Security (000004.SZ), Shengtong Stock (002599.SZ) and other trading boards, while Zhongqingbao (300052.SZ) rose 14%.

Multimodal AI Index Sections

Tom Cat stated in the first quarter of 2024 research report that the company is advancing the construction of artificial intelligence vertical models and computing infrastructure in an orderly manner. At present, the company and West Lake Xinchen jointly developed the "Tom Cat Emotional Companion Model" is in the process of further research and development, the model will effectively help the company's Tom Cat AI products to achieve personalized, high IQ, high EQ companion interaction. In addition, the first phase of the intensive computing center built by the company in Dingxi, Gansu Province is planned to be completed in the first half of the year. After completion, the computing center will provide basic services such as computing power storage for the company's artificial intelligence business. The construction of the emotional companion vertical model and computing infrastructure will provide strong support for the long-term development of the company's artificial intelligence business.

Aofei Entertainment said on the interactive platform that it currently holds an unlimited 5% stake in Lightyear and has its own AI dialogue product, Turing Robot Open Platform. Lightyear Infinite is mainly engaged in the research and development and commercial application of robot artificial intelligence and robot operating system, and has leading advantages in semantic understanding, machine vision, multi-modal human-computer interaction, deep learning, robotics and other fields. At present, the company on-line AI dialogue small program "happy and gray wolf plus", positioning children users, to provide parent-child companionship experience.

Guoxin Securities pointed out that the fourth productivity revolution represented by AI has come and become the core driving force driving the performance of the media sector, and the investment opportunities of the AIGC application-driven sector are worth looking forward.

since this year, the global AI model has been continuously upgraded iteratively, including overseas Sora, Llama3, etc., domestic Kimi, Kunlun tiangong AI, step star, etc. According to the Securities Times-Data Treasure statistics, multi-modal AI concept stocks in 2023 research and development spending totaled 32.753 billion yuan, accounting for 11.2 percent of revenue, which is 4.46 times the overall level of A shares in the same period.

In addition, this year is also expected to be the first year of the outbreak of AI applications, and the global technology industry is ready to move.

in may, the global AI field ushered in an important month of activities, with blockbuster releases throughout the month. In addition to the GPT-4o release, Ali, Tencent, byte beat, Google, Microsoft and other domestic and foreign technology manufacturers have action, the degree of excitement can be called "AI month".

On May 9, Aliyun officially released the Tongyi Thousand Questions 2.5 large model and opened source Tongyi Thousand Questions 110 billion parameter model.

on may 14, Tencent announced that its big model of mixed yuan Wen sheng tu has been fully upgraded and opened to the outside world. currently, it has been released on Hugging Face platform and Github. it includes complete models such as model weight, reasoning code and model algorithm, which can be used for free commercial use by enterprises and individual developers.

according to the official micro news of volcano engine today, byte beat will hold the "2024 spring volcano engine FORCE motive power conference" on may 15. this conference focuses on AI theme, takes large model application as the core and AI landing as the guide, shows the practical application of volcano engine in the fields of large model and cloud computing, and joins hands with experts and enterprise technology leaders in the fields of automobile, mobile phone terminal, finance, consumption, internet, etc, discuss together the transformation opportunities in the AI era.

Google announced that the 2024 I/O Developer Conference will be held at 13: 00 Eastern Time on May 14. The focus of this conference will be artificial intelligence. The conference will also hold more than 10 "technical sub-forums" for developers related to generative artificial intelligence, with topics including the latest progress of Gemma, a large model of artificial intelligence, and learning how to use Gemini for multimodal retrieval to enhance generation. At this I/O developer conference, Google will also release the new Android 15 mobile operating system. It is expected that Android 15 will bring many new artificial intelligence functions. Google may even release a new large language model for Android 15.