AI (Artificial Intelligence) has become a keyword in the global technology venture capital industry over the past year, and the future may only become more lively.
In the past week, global technology giants have intensively released the latest AI products and technologies. On May 13th, OpenAI snatched attention with GPT-4o. One day later, Google also released the AI product FamilyBucket, which left stunning memories with the Veo video model, Project Astra, and the new version of AI search. Multimodal fusion technology has become a common focus of both parties, both aimed at advancing towards the goal of General Artificial Intelligence (AGI).
On the other side of the ocean, ByteDance released the Doubao big model family, and Tencent finally handed over the answers to the agent product "Tencent Meta" and the big model assistant APP. Li Kaifu, the creator of Zero One Everything, made his official debut with the Yi Large closed source model, which has a parameter of 100 billion. He also unveiled the strategic layout of the open source closed source dual track model.
In the near future, there will be a heavyweight conference for two tech giants: on May 21st, Microsoft will hold the Build annual developer conference; On June 10th, Apple will hold a global developer conference.
Several industry insiders stated in an interview with China Business Daily that the artificial intelligence race is rapidly exploding and becoming a global focus of attention. The AI competition among technology giants will continue to heat up in the future. Driven by the strong demand for AI, major manufacturers are concentrating their efforts to accelerate their layout. At present, technology companies such as Google, Microsoft, Amazon, Apple, and Meta have begun to heavily invest in the field of artificial intelligence, continuously increasing their capital expenditures.
Technology giants working together
After releasing two popular models, ChatGPT and Sora, OpenAI recently launched its latest achievement, GPT-4o.
The release of GPT-4o has sparked widespread discussion in the AI community. The industry believes that the stunning feature of GPT-4o lies in two aspects: firstly, it reduces the delay of voice interaction to 300ms; Next is the end-to-end multimodal native large model. The understanding ability of voice assistants has also made a qualitative leap, not only analyzing and discussing the images or videos they see, but also recognizing different emotions when users speak.
Just one day later, Google released its AI product family bucket at the I/O 2024 Developer Conference, covering everything from search to Gmail, TPU, voice assistant Astra, and multimodal video model Veo. In the 110 minute speech, Google CEO Pichaiti and AI made 121 appearances.
At the same time, across the ocean in China, the family of ByteDance self researched big model Doubao big model (the original skylark big model), with nine models, debuted for the first time. The inference price of the Doubao big model has become a major highlight, with its main model priced at only 0.0008 yuan/thousand tokens in the enterprise market, which can process over 1500 Chinese characters at 0.8 millimeters.
On May 17th, Tencent announced the open source of its hybrid Wenshengtu model, which has been released on the Hugging Face platform and Github. It includes complete models such as model weights, inference code, and model algorithms, and is available for free commercial use by both enterprises and individual developers. Tencent has also announced an upgrade to the hybrid model, launching three versions of the model with different quality and cost characteristics. It already has over 600 business integrated into the model internally.
After nearly a year of continuous "frenzy", OneThing has recently officially released the Yi Large with a parameter scale of 100 billion. Li Kaifu also announced the global launch of an API open platform that empowers the 2B ecosystem.
In the coming period, AI will still be the main protagonist.
More than half of the session directory related to "AI" will be held at the Microsoft Build Annual Developer Conference on May 21st. Copilot will be one of the key topics of the annual Microsoft Build Developer Conference. Microsoft recently announced the integration of artificial intelligence technology into the development environment.
The World Developers Conference (WWDC24) will be held from June 10th to 14th local time, and Apple hopes to catch up with its competitors in artificial intelligence. Industry insiders believe that iOS18 will be a "relatively breakthrough" software update, and artificial intelligence is expected to be integrated into the entire operating system. Industry insiders predict that Apple may launch a brand new upgraded Siri voice assistant, which introduces a new generative AI system into its underlying technology.
The latest news suggests that Apple may collaborate with OpenAI to use external AI in the iOS system, while also negotiating with Google to introduce Gemini.
Looking at market value in the short term and applications in the long term
On the evening of May 14, the two Internet giants Alibaba and Tencent released their financial reports almost simultaneously. The financial reports show that both giants are gradually finding the driving force for a new round of growth: artificial intelligence. Through a new round of large-scale investment in artificial intelligence and support for various businesses, the Internet giants are looking forward to gathering more potential energy and leading the industry back to the rising channel.
Multiple institutions have reported that in the medium to long term, the company is actively laying out the development of AI technology, with a diversified product matrix corresponding to broad potential application directions. Technological progress may become an important growth multiplier in the medium to long term.
However, the new wave of AI technology is just beginning. In addition to a few giants such as Nvidia and AMD, which have directly benefited from the "AI era shovel sellers", Internet technology giants are still in the stage of large-scale investment "burning money".
According to Tencent, the R&D expenditure in the first quarter of 2024 reached 15.6 billion yuan, and the R&D investment in six years from 2018 to present has exceeded 285.3 billion yuan. On the Alibaba side, not only has Taotian established an AI e-commerce team internally, but the international station has also released full chain AI foreign trade products. Alibaba Cloud AI has also invested in the product matrix of cloud computing, especially AI infrastructure. In addition to focusing on the Tongyi Qianwen model, Alibaba has also completed investments in five leading Chinese models, namely Unicorn (Moon Darkness, MiniMax, Zhipu, Baichuan, and Zero One Universe).
Relatively speaking, American giants are all in AI.
The financial report shows that Microsoft's capital expenditure in the second quarter of the 2024 fiscal year was $11.5 billion, reaching a new high in seven years. Google's capital expenditures are also rapidly expanding. In the fourth quarter of 2023, Google's capital expenditure was $11 billion, a year-on-year increase of 45.1%. The company revealed that its capital expenditure will continue to increase in 2024 to support sustained investment in AI.
Meta and Apple have also expressed their emphasis and expectations for the future of artificial intelligence business. For example, in order to accelerate the development of AI products, it is planned to increase the capital expenditure for 2024 from the previously predicted $30-37 billion to $35-40 billion.
Compared to the rapidly increasing investment, the bonus of AI dividends on performance has not yet been clearly reflected in the latest financial reports of various giants. For example, people have high expectations for Microsoft's AI business that has invested in OpenAI. However, Microsoft's flagship cloud service product Azure in the latest fiscal quarter continued the growth trend of the previous quarter, but after excluding the impact of exchange rates, the growth rate was basically the same as the previous quarter and did not show the expected high growth.
IDC China Research Director Lu Yanxia stated that the impact of AI on corporate performance is currently not significant, and the application of large models is not yet mature. Some sub areas that can be applied will definitely be helpful for improving corporate performance, but the initial huge cost investment will actually slow down performance and bring financial pressure to a certain extent.
Will the future AI craze cool down after the frenzy? Lu Yanxia believes that it depends on whether large-scale AI models can be widely applied. If it cannot be widely and on a large scale, it may lead to cooling or even cooling.
At present, the significance of AI to giants lies more in the enthusiasm of the capital market and its impact on the stock market.
According to relevant data, in 2023, AI chip giant Nvidia rose nearly fivefold, with a market value climbing to $1.8 trillion, becoming the third largest technology company in the world by market value; ARM and AMD have tripled, while Intel has increased by over 58%; Microsoft also invested in ChatGPT, and its stock nearly doubled, surpassing Apple to become the world's largest technology giant by market value.
As the initiator of this AI boom, OpenAI's valuation has soared rapidly, and now it has become the world's third largest technology unicorn enterprise after ByteDance and SpaceX.
It can be imagined that artificial intelligence is one of the main investment themes for 2024, whether from a policy perspective, industry trends, or enterprise level.
However, an AI industry practitioner told reporters that after the boom in artificial intelligence concept stocks last year, investors will become more rational and cautious this year, and the true winners of artificial intelligence may be distinguished from those like C3. ai who are just filling in the numbers.
Chinese AI needs to catch up
Compared to All in AI, a technology giant on the other side of the ocean, although China has quickly launched a hundred model war, there is still some "thunder, rain, and little". Because last year, only the United States saw an increase in investment and financing for AI startups, while Europe and Asia saw a decline.
According to CB Insights data, in 2023, the number of investment and financing for American AI startups reached 1151, with a total financing amount of $31 billion, a year-on-year increase of 14%, and attracted 46% of all AI transactions worldwide. In 2023, the number of investment and financing in the domestic AI field was about 232, a year-on-year decrease of 38%; The total financing amount is about 2 billion US dollars, a year-on-year decrease of 70%. This means that although major domestic giants have launched self-developed models, overall AI investment has still declined.
"Fortunately, in the AI era, China has its own unique talent advantage. On the one hand, the development of China's Internet industry is second only to the United States, and China's Internet giants have long-term investment in artificial intelligence, with talent and technology accumulation." Yang Fei, an industrial economic observer, told reporters that, on the other hand, among the leading AI enterprises in the United States, the proportion of Chinese is very high. These people have close ties with China, and will have the opportunity to attract them back to work for the country in the future.
For example, a quarter of the engineers in OpenAI's ChatGPT R&D team are Chinese, and 4 out of 13 members in Sora's R&D team are Chinese; The CEOs of global AI chip industry leaders Nvidia and AMD are both Chinese, and among well-known scientists in the field of AI in the United States, Chinese are abundant.
Yang Fei believes that China's advantage lies in its leading position in improving enterprise productivity through the application of AI and high-speed broadband. As long as breakthroughs are made in technology, China will evolve faster in terms of AI applications than the United States. However, China also faces resource constraints on chips and computing power. At present, the OpenAI team is trying to connect 10 million GPUs together, while each of the domestic Internet giants has only thousands or tens of thousands of GPUs. "China should coordinate Internet enterprises to focus on limited GPUs for AI product research and development at the national level, and seize the time to let the leading enterprises take the lead to realize the mass production of China's independent high-performance AI GPUs." Yang Fei said.
In response to the current chaotic situation in the domestic big model market, Li Kaifu called on entrepreneurs to pay attention to TC-PMF (Technology Cost Product Market Fit), reject the ofo style money burning strategy, and enable big models to run long distances with a healthy and healthy ROI (Return on Investment), heading towards the AI 2.0 transformation belonging to China. Li Kaifu believes that AI inclusive access to the Chinese market is inevitable.
The battle for AI dominance among technology giants has already begun, and a new round of competition will only become more intense and comprehensive. The future is still full of suspense.