At the Artificial Intelligence Big Model Technology Summit Forum held on April 8th, Tian Qi, the chief scientist in the field of artificial intelligence at Huawei Cloud, stated that big models are a bridge connecting the technology ecosystem and the business ecosystem, and are the core of the future AI ecosystem. Under the driving effect of the AI big model on the industrial chain, many listed companies have recently updated their AI layout progress. Artificial intelligence has become a new "traffic gateway" for listed companies. At the same time as announcing its entry into AI, the company's stock price also rose. According to Wind data, during the AI market boom since the beginning of this year, the AI index has risen by over 60%, the AIGC concept index has risen by 62%, the ChatGPT concept index has risen by 86%, and the AI computing power concept index has risen by 78%. From a long-term perspective, AI is expected to form a significant industry trend, with a group of high-quality companies achieving dual expansion in profitability and market value. However, with the rapid increase in trading heat, the volatility of AIGC market may further expand, and it is still necessary to observe the performance of fundamentals AI big model is the most popular AI subdivision field in recent years. Compared with small models, which have limited data, bottleneck capabilities, severe customization and fragmentation, and lack of large-scale replication and emergence capabilities, AI big models have many advantages such as common, generalized, and large-scale replication of multiple scenarios, and are considered to realize AGI (General Artificial Intelligence, Artifical

The important direction of General Intelligence. In fact, the AI big model is more like a synthesizer of the development of AI to a certain stage, integrating various AI technologies that have emerged over the years, supplemented by excellent data and sufficient computing power, gradually demonstrating amazing and powerful capabilities. At the Boao Forum for Asia, experts generally believed that the next few years would usher in a dividend period for AI technology, and the AI development speed was far faster than people had expected. ChatGPT is a milestone event in the field of artificial intelligence. The ability demonstrated by pre trained large models indicates that universal artificial intelligence is not unreachable. Pre trained large models can be replicated and commercialized, and the wave of artificial intelligence it brings will not only change the way human-computer interaction is conducted, but also greatly enhance high productivity. "Chief scientist in the field of Huawei Cloud Artificial Intelligence, academician of the International Eurasian Academy of Sciences, IEEE

According to Fellow Tianqi. The reliability and security of AI must be an area that must be taken seriously in the present and future. Tian Qi believes that "firstly, the country needs to introduce relevant policies as soon as possible to prevent the abuse of AI products; secondly, it is necessary to ensure the controllability and reliability of AI products at the technical level. Issues such as cultural images and Q&A may involve ethical issues, and it is necessary to pay attention to data privacy and security and improve the interpretability of the model." When the driving effect of AI models on social productivity has become increasingly apparent, we have no reason to give up eating for fear of choking. In the long run, AI has entered multiple industries and has pressed the accelerator button. Utilizing AI to reduce costs, increase efficiency, and liberate productivity has become a trend On April 8th, the Chief Scientist of Huawei Cloud Artificial Intelligence, Tian Qi, will introduce the "Progress and Applications of Pangu Big Model" at the Artificial Intelligence Big Model Technology Summit Forum. According to the official website of Huawei Cloud, the "Pangu series AI big models" that Huawei is about to launch are NLP big model, CV big model, and scientific computing big model (meteorological big model). The NLP model is considered the AI model that is closest to human Chinese comprehension ability, while the CV model for the first time combines image discrimination and generation capabilities. On April 7th, Alibaba Cloud announced that it would invite users to test the self-developed model "Tongyi Qianwen". Ali Dharma Institute started the research and development of large model in 2019, and has accumulated many years of technology in NLP natural language processing and other aspects. According to industry insiders, Shangtang Technology, Alibaba Big Model products, and Tonghuashun AI related products are expected to be released in the following week. Tencent and ByteDance will also release new AI models in succession. Following Baidu's launch of the "ERNIE Bot" chat robot product, the domestic Internet company's AI model product ushered in a hundred flowers bloom. | AI

Will for Industries be the next explosive point? Currently, the significant progress in AI models in terms of capabilities has made it possible to liberate productivity and creativity at the individual level. Therefore, major internet companies are embracing the AI model and reshaping various products and services based on their existing business scenarios. For example, combining GPT technology with search engines can bring more precise and better user experience; The integration of intelligent customer service in e-commerce with GPT technology can provide a customer service experience that is more in line with the logic of human dialogue; By combining GPT technology with office software and conference software, efficiency can be significantly improved, and so on. Next, from the perspective of industrial transformation and upgrading, AIfor.

Industries will become the next explosive point of artificial intelligence, and the integration of AI with various industries will go deeper. For example, in the field of vertical finance, Bloomberg has released Bloomberg GPT, which provides services to financial analysts based on massive financial data, marking the beginning of the deep integration of universal big models with the financial industry. So, big models are seen as an important path to solving various AI challenges, and how should they truly support AI in the future

for Industries？ Specifically, the connection between large models and industries means that large models need to truly reduce usage barriers and structural costs, making them convenient for users in various industries. The big model will become the operating system of AI. From multiple perspectives of usage and cost, the big model will accelerate its adaptation to hardware, and all AI algorithms above can be built and applied around the big model, "Tian Qi judged Under the background of widespread attention to ChatGPT and domestic AI models, more and more listed companies are optimistic about the future development opportunities of the AI track, as AI layout presses the accelerator button and is wary of investment risks. The rapid development and application of ChatGPT is expected to further drive the incremental demand for domestic AI servers. In the medium to long term, there will be an increase in demand for computing power, large capacity and high bandwidth storage, and high-speed interconnect chips, which will promote the company's general AIBOX, AI servers, and industry AI terminals. From basic layer fields such as AI computing power and data, to application fields such as smart home and smart healthcare, listed companies are accelerating their layout based on their own business characteristics. Since the beginning of this year, the emergence of AI concepts has ignited investors' investment enthusiasm. As of the latest closing date, the stock prices of Cambrian-U, Yuncong Technology-UW, Haitian Ruisheng, and Kunlun Wanwei have increased by more than 2.5 times this year, while the stock prices of companies such as 360, Toulouse, Wanxing Technology, and Zhonghua Online have risen by 184%, 158%, 149%, and 142% respectively. With the release of AI products and technology catalysis, the current AI market may continue. But with the rapid increase in trading heat, the volatility may further expand. AI enters the core production system of enterprises and begins to create greater value. The penetration rate of the artificial intelligence industry has accelerated and will reach 20% by 2026. The development of artificial intelligence faces two major trends: firstly, from small models to large models, the computational power demand of AI algorithms has increased by 400000 times in the past 10 years. Large models may include highly customized small models, leading to market concentration towards large companies: high capital threshold, large model development and training at a time of 12 million US dollars; High technical threshold, deep optimization of AI framework and strong parallel computing ability. There may be a 0-15% upward potential for the AIGC index in the future, and the maximum retracement point may correspond to a current index decline of around 8% -30%. Whether AIGC can evolve into a major theme market in the future still requires observation of the fundamental cashing situation.