On December 30th, Science and Technology Innovation Board Daily reported that competition for AI chips is becoming increasingly fierce. According to media reports, Nvidia, as a computing power giant, is betting on robot technology, hoping it can become the company's next important growth driver.
The ChatGPT moment for physical artificial intelligence and robotics technology is coming soon, "said Deepu Talla, Vice President of Robotics at NVIDIA.
Currently, robots, especially humanoid robots, are still in the early stages of scaling up, and the industry is still committed to reducing costs, improving product accuracy, and expanding scale.
Tella pointed out that the driving force behind the transformation of the robotics market mainly comes from two technological breakthroughs: the explosive growth of AI models and the ability to train robots on these models through simulated environments. The latter is an extremely important development that helps bridge the "simulation to reality gap" and ensures that robots trained in virtual environments can operate effectively in the real world.
In the past 12 months, this gap has matured enough (gradually narrowing). Now we can combine generative AI for simulation experiments, which was not possible two years ago
In the field of robotics, Nvidia positions itself as a supplier of robotics technology platforms, providing full stack solutions covering various layers of software and robot chips for training robots. The company plans to launch the latest generation of small computer Jetson Thor for humanoid robots in the first half of 2025.
At the same time, Goldman Sachs has significantly raised its expectations for the size of the humanoid robot market in its latest research report. By 2035, the market size of the humanoid robot industry will reach $38 billion, more than six times the previously predicted value of $6 billion.
When explaining the reasons for raising expectations, Goldman Sachs, like Nvidia Deepu Talla, also pointed out the important role of AI models in robot training. The accelerated evolution of AI, technological breakthroughs, and increased capital expenditure investment are the core driving factors that have prompted us to adjust our forecasts. At present, significant progress has been made in end-to-end AI, and models can self train through this AI without the need for human engineers to manually write all the code. This accelerates the development of robots, enabling these devices to complete more tasks faster and adapt to new situations (such as working outside factories)
Goldman Sachs predicts that humanoid robots are expected to be first applied in factories between 2024 and 2027, while entering the consumer market is expected between 2028 and 2031.
Overall, many companies at home and abroad have accelerated their deployment of robots recently.
OpenAI is considering independently developing humanoid robots after investing in Figure and 1X, and the company has restarted its robot team; Guangzhou Automobile Group releases its embodied intelligent humanoid robot GoMate; The Leju humanoid robot production line has officially started, with an expected annual output of 200 humanoid robots; Shanghai Matrix Super Intelligent Integrated System Co., Ltd. has released its self-developed full-size universal humanoid robot prototype MATRIX-1; Uniqlo Technology releases new universal humanoid robot Wanda 2.0
Great Wall Securities pointed out that the current humanoid robot industry chain is in the process of transitioning from; quot; 0-1& quot; Commercialization landing& quot; 1& quot; In the accelerated promotion stage, it has been evident since the beginning of this year that there has been a significant increase in the number of entrants, and the entry of large companies has become a clear trend. Analysts are optimistic that humanoid robots, as one of the important directions for the landing of general artificial intelligence, the forefront direction of new quality productivity, and an important branch of the first economy, will continue to accelerate their commercialization process driven by policies, and are expected to welcome the arrival of the first year of mass production in 2025.