（来源：英伟达）① Nvidia predicts that artificial intelligence will have a significant impact of up to $20 trillion in all industries using the technology; ② Starting from Tuesday, some companies have started showcasing the new B200 chip servers, marking a new milestone in AI computing power hardware.
On October 9th, Finance and Economics Society news (Editor: Shi Zhengcheng) local time Tuesday, global computing power leader Nvidia kicked off the 'AI Summit' in Washington D.C., USA. It is reported that during the three-day event, the company will focus on demonstrating to policymakers and the media how Nvidia's chip system and customizable AI services can benefit the entire society.
The main figure leading the activities on Tuesday was not the familiar 'Leather Jacket Old Huang' to investors, but Nvidia's Vice President and General Manager of Enterprise Platforms, Bob Pette. This indicates that the entire event is more focused on AI applications rather than new product releases.
(Source: NVIDIA)
Pette stated that the world is currently on the edge of seeing more industries adopting AI, with expectations for artificial intelligence to have a significant impact of up to $20 trillion in all industries using this technology.
In the keynote speech on Tuesday, Pette elaborated on the 'next stage of artificial intelligence', ranging from advanced AI agents to factories operated by robots, weather forecasting, cancer treatment, and exploring extraterrestrial life.
Through thousands of pre-trained 'AI agents' (Nvidia NIM agent), industrial, scientific, and medical institutions worldwide are undergoing a 'industrial revolution-level' transformation of productivity. For example, the National Cancer Institute in the USA is utilizing Nvidia's medical AI services for medical image analysis and extracting information from large databases. At the same time, Nvidia's AI is aiding pharmaceutical companies and researchers in screening new drug molecules, significantly reducing the time needed to develop new drugs.
Similar cases of AI agents include the cybersecurity AI Blueprint Framework, which can monitor and respond to cybersecurity vulnerabilities and threats in the entire system in real time, as well as the space exploration AI that detects weak radio signals from space in real time. The 'Digital Earth Twin' created by AI - NVIDIA Earth 2 can help meteorologists predict natural disasters more accurately and quickly.
Petzinger stated that in NVIDIA's CUDA library, there are now over 4,000 accelerated applications, helping various industries achieve breakthroughs.
According to NVIDIA, the next wave of AI will be 'Physical AI', referring to robots interacting in the physical world, with the machines themselves controlled by remote computers.
Ultimately, these applications still rely on NVIDIA's computing devices to operate. As we enter the last quarter of 2024, the highly anticipated GB200 AI server is finally ready for shipment.
Microsoft Azure revealed on Tuesday that the company has received servers equipped with GB200 chips, making it the first cloud computing service provider in the world to adopt the Blackwell system.
OpenAI also announced the delivery from NVIDIA, with developers already using the first batch of DGX B200 engineering machines.
Nvidia had previously anticipated that in response to the market's frenzy for machines, they expected to ship the latest AI servers worth "tens of billions of dollars" in the fourth quarter.
After the end of this week's event, Nvidia will continue to hold an "AI Summit" in Mumbai, India at the end of October, and then move to Tokyo, Japan in November. Both of these countries have government-supported large-scale computational infrastructure investment plans.