News from the AI & ML world
DeeperML - #aichips
|
@www.artificialintelligence-news.com
//
Nvidia is actively working to maintain its presence in the rapidly expanding Chinese AI chip market, even amidst tightening U.S. export controls and rising domestic competition from companies like Huawei. To navigate these challenges, Nvidia is developing a cheaper, stripped-down AI chip based on its Blackwell architecture, specifically tailored for the Chinese market. This new chip is designed to comply with U.S. regulations while offering competitive performance, aiming to regain some of the market share Nvidia has been losing.
This strategic gamble involves significant cost reductions and performance trade-offs. The new processor, expected to be priced between $6,500 and $8,000, will use the RTX Pro 6000D foundation with GDDR7 memory, forgoing the high-bandwidth memory found in premium variants and Taiwan Semiconductor's advanced CoWoS packaging technology. This decision is driven by the need to meet export restrictions and lower manufacturing complexity, following a $5.5 billion writedown Nvidia absorbed after the prohibition of its H20 sales in China. Huawei's growing influence in the Chinese AI chip landscape is a key factor driving Nvidia's strategy. Huawei's Ascend 910C and 910B processors are gaining adoption among major domestic tech firms, challenging Nvidia's dominance, particularly in inference applications. The competitive pressure extends to complete infrastructure solutions, with Huawei's CloudMatrix 384 rack system directly challenging Nvidia's Blackwell GB200 NVL72 configuration. The company hopes the new budget option will help them survive the growth of the internal competion, Recommended read:
References :
@thetechbasic.com
//
References:
thetechbasic.com
, Bloomberg Technology
,
Nvidia CEO Jensen Huang has denied allegations of AI chip diversion to China, stating there is no evidence of any illegal flow of hardware. Speaking in Taipei, Huang addressed global concerns regarding the transfer of advanced technology to restricted markets. He emphasized the impracticality of covertly shipping Nvidia’s large and complex systems, such as the Grace Blackwell, which weighs nearly two tons and comprises numerous high-powered GPUs and processors.
Huang highlighted the strict oversight by Nvidia's customers, which include major tech giants like Microsoft, Amazon, Alphabet, and Meta Platforms. These customers are fully aware of and compliant with US export controls, as losing access to Nvidia's technology would be a significant setback for them. The CEO underscored the importance of trust and transparency in Nvidia's supply chain, emphasizing the value customers place on maintaining access to Nvidia's cutting-edge AI solutions. In other news, Nvidia is reportedly planning to establish a new research and development center in Shanghai to further solidify its position in the AI chip market. This initiative aims to tailor solutions specifically for Chinese clients while also contributing to global R&D efforts in areas such as chip design verification, product optimization, and autonomous driving technologies. The Shanghai government has reportedly expressed initial support for this project, which would expand Nvidia's presence in the city beyond its current sales and support operations. Recommended read:
References :
@blogs.nvidia.com
//
Nvidia is currently facing pressure from the U.S. government regarding AI GPU export rules. CEO Jensen Huang has been advocating for the Trump administration to relax these restrictions, arguing they hinder American companies' ability to compete in the global market. Huang stated at the Hill and Valley Forum that China is not far behind the U.S. in AI capabilities, emphasizing the need to accelerate the diffusion of American AI technology worldwide. He also acknowledged Huawei's progress in computing, networking, and software, noting their development of the CloudMatrix 384 system. This system, powered by Ascend 910C accelerators, is considered competitive with Nvidia's GB200 NVL72, signaling the emergence of domestic alternatives in China.
Despite Nvidia's pleas, the Trump administration is considering tighter controls on AI GPU exports. The administration plans to use chip access as leverage in trade negotiations with other nations. This approach contrasts with Nvidia's view that restricting exports will only fuel the development of competing hardware and software in countries like China. According to the AI Diffusion framework, access to advanced AI chips like Nvidia’s H100 is only unrestricted for companies based in the U.S. and "Tier 1" nations, while those in "Tier 2" nations face annual limits and "Tier 3" countries are effectively barred. Adding to the complexity, Nvidia is also engaged in a public dispute with AI startup Anthropic over the export restrictions. Anthropic has endorsed the Biden-era "AI Diffusion Rule" and has claimed there has been chip smuggling to China. An Nvidia spokesperson dismissed Anthropic's claims about chip smuggling tactics as "tall tales," arguing that American firms should focus on innovation instead of trying to manipulate policy for competitive advantage. As the May 15th export controls deadline approaches, the tensions continue to rise within the AI industry over the balance between national security, economic prosperity, and global competitiveness. Recommended read:
References :
Ali Azhar@AIwire
//
Nvidia CEO Jensen Huang has expressed concerns about the growing competition from Huawei in the AI chip market, a notable admission highlighting the shifting dynamics within the global tech landscape. Geopolitical tensions and tightening U.S. export controls are reshaping technology supply chains, creating both challenges and opportunities for companies worldwide. Huang has previously called for the Trump administration to relax AI GPU export restrictions to support U.S. industry; however, the administration is considering stricter controls as leverage in trade negotiations.
Huawei is actively developing its Ascend series to challenge Nvidia's dominance. The new Huawei Ascend 910D AI processor is designed to compete directly with Nvidia's Blackwell and Rubin GPUs. Huawei plans to begin testing the Ascend 910D processor, aiming for it to surpass the performance of Nvidia's H100. To achieve H100 performance levels, Huawei will have to redesign the internal architecture of the Ascend 910D and possibly increase the number of compute chiplets. The company has approached local Chinese firms to evaluate the chip's performance, with initial samples expected by late May 2025. In response to the evolving AI landscape, Nvidia is also focusing on strengthening its cloud infrastructure. NVIDIA Blackwell GPUs are now being deployed on NVIDIA DGX Cloud and Oracle Cloud Infrastructure (OCI) to develop and run reasoning models and AI agents. Oracle Cloud Infrastructure is deploying NVIDIA Blackwell GPUs on NVIDIA DGX Cloud and Oracle Cloud Infrastructure (OCI) to develop and run reasoning models and AI agents. This move is intended to solidify Nvidia's position and provide its customers with advanced AI capabilities through cloud-based solutions. Recommended read:
References :
@thequantuminsider.com
//
References:
www.cnbc.com
, the-decoder.com
,
China is making significant strides in artificial intelligence, quantum computing, and robotics, signaling a growing technological prowess. Huawei is reportedly developing new AI chips, the Ascend 910D, aiming to rival Nvidia's high-end processors. These chips are undergoing testing with Chinese tech companies, and early samples are expected as soon as late May. This move aims to diminish reliance on foreign technology and support China's growing AI ambitions, especially given the restrictions placed on sales of advanced Nvidia products to China.
Chinese scientists have also made advancements in quantum technology with the development of a drone-mounted quantum sensor capable of detecting submarines. This sensor, utilizing a Coherent Population Trapping atomic magnetometer, addresses limitations of traditional magnetic anomaly detectors in low-latitude regions. Offshore trials near Weihai, Shandong province, demonstrated the system's high accuracy and stability in detecting minute changes in Earth's magnetic field, presenting possibilities for underwater resource mapping and military applications. Further testing in harsher conditions is needed before operational deployment. President Xi Jinping has emphasized the need for self-reliance in AI development. At a Politburo meeting, he called for increased support in research, talent training, and intellectual property protection. He also urged for strengthening basic research and key technologies like high-performance chips and basic software. Additionally, he pushed for expediting AI laws and regulations to ensure safety and control, highlighting China's commitment to closing the AI gap through homegrown innovation. Simultaneously, in the realm of robotics, China is emerging as a serious contender, demonstrated by the recent humanoid robot half-marathon, showcasing progress in humanoid robot technology and the country's manufacturing and engineering capabilities. Recommended read:
References :
Harsh Sharma@TechDator
//
Huawei is intensifying its challenge to Nvidia in the Chinese AI market by preparing to ship its Ascend 910C AI chips in large volumes. This move comes at a crucial time as Chinese tech firms are actively seeking domestic alternatives to Nvidia's H20 chip, which is now subject to U.S. export restrictions. The Ascend 910C aims to bolster China's tech independence, providing a homegrown solution amidst limited access to foreign chips. The chip combines two 910B processors into one package, utilizing advanced integration to rival the performance of Nvidia’s H100.
Huawei's strategy involves a multi-pronged approach. Late last year, the company sent Ascend 910C samples to Chinese tech firms and began taking early orders. Deliveries have already started, signaling Huawei's readiness to scale up production. While the 910C may not surpass Nvidia's newer B200, it is designed to meet the needs of Chinese developers who are restricted from accessing foreign options. The production of the Ascend 910C involves a complex supply chain, with parts crafted by China's Semiconductor Manufacturing International Corporation (SMIC) using its N+2 7nm process. Despite the challenges from Huawei, Nvidia remains committed to the Chinese market. Nvidia is reportedly collaborating with DeepSeek, a local AI leader, to develop chips within China using domestic factories and materials. This plan includes establishing research teams in China and utilizing SMIC, along with local memory makers and packaging partners, to produce China-specific chips. CEO Jensen Huang has affirmed that Nvidia will continue to make significant efforts to optimize its products to comply with regulations and serve Chinese companies, even amidst ongoing trade tensions and tariffs. Recommended read:
References :
@www.pcguide.com
//
Huawei is emerging as a significant competitor to NVIDIA in the artificial intelligence hardware arena. The Chinese tech giant recently showcased its CloudMatrix 384 supercomputer at the Huawei Cloud Ecosystem Conference 2025, boasting superior performance metrics compared to NVIDIA's offerings. According to Huawei's CEO of Cloud Computing, Zhang Ping’an, the CloudMatrix 384 stands out for its "high density," "high speed," and "high efficiency," leading in computing power, interconnect bandwidth, and memory bandwidth. This development signals a notable shift in the AI chip sector, especially given the limitations US restrictions place on NVIDIA's exports to China.
The CloudMatrix 384 leverages the upcoming Ascend 910C chips, built on a 7nm process node. According to sources on Twitter, the supercomputer is better than the NVIDIA GB300 NVL72 on all metrics. This technological advancement could have a significant impact on the future of AI development and deployment across various industries. The rivalry between Huawei and NVIDIA is expected to intensify, reshaping the competitive landscape of the AI hardware market. Simultaneously, cybersecurity concerns are being raised by the Canadian Centre for Cyber Security regarding Chinese state-sponsored actors targeting network edge routers. The agency has observed increasing activity, including that associated with SALT TYPHOON, aimed at compromising these devices across critical infrastructure sectors. Threat actors exploit vulnerabilities in security and network edge routing devices to enter a network, monitor, modify, and exfiltrate network traffic, or potentially move deeper into victim networks. The Cyber Centre urges organizations to bolster awareness and leverage guidance to protect their networks, including applying patches and adhering to manufacturer's security guidance. Recommended read:
References :
ashilov@gmail.com (Anton@tomshardware.com
//
Nvidia is facing a significant financial setback due to tightened U.S. export restrictions on its H20 AI chips to China. The U.S. government, citing concerns about the chips' potential use in Chinese supercomputers, now requires Nvidia to obtain special licenses for all H20 chip exports to China, Hong Kong, and Macau. This indefinite restriction, revealed on April 9, 2025, has forced Nvidia to write off $5.5 billion in GPUs, as unsold inventory and sales commitments take a hit. The H20 chip, a modified version of the H100 designed to comply with previous export rules, was a key product for Nvidia in the Chinese market, powering AI applications like chatbots and image generators.
This move by the U.S. government is seen as a further escalation in the ongoing tech rivalry between the two nations. The restrictions aim to prevent the H20 chips, known for their memory and interconnect bandwidth, from contributing to China's AI ambitions, particularly in supercomputing. While Nvidia can still apply for licenses, the expectation is that denials will be the norm, effectively cutting off a significant revenue stream. This stricter stance represents a shift from previous policies, which permitted the export of the H20 as a less powerful alternative to other restricted chips. Even AMD's Instinct MI308 is reportedly facing export restrictions too. The financial impact of these restrictions is substantial. Nvidia anticipates a $5.5 billion charge in its first fiscal quarter of 2026, which ends on April 27, 2025, due to inventory, purchase commitments, and related reserves. This news sent Nvidia shares tumbling in after-hours trading, raising concerns about the company's future prospects in the crucial Chinese market. The regulatory filing indicated that the new controls are in place to prevent the chips from finding their way into Chinese supercomputers. Recommended read:
References :
Sean Hollister@The Verge
//
Nvidia CEO Jensen Huang unveiled the company's next generation of AI chips at the GTC 2025 conference, including the Blackwell Ultra GB300 and Vera Rubin, its next AI superchips. During the conference, Huang emphasized the advancements in AI and his predictions for the industry's future, noting that AI is at an "inflection point" and highlighting the evolution of AI from perception to generative AI, and now to agentic AI which can understand context and generate answers. Nvidia's new roadmap includes the Blackwell Ultra, slated for release in the second half of 2025, and the Vera Rubin AI chip, expected in late 2026.
The Blackwell Ultra isn't built on a completely new architecture but it offers enhanced capabilities, including 20 petaflops of AI performance and 288GB of HBM3e memory. In addition to the chip announcements, Nvidia revealed that it is building the Nvidia Accelerated Quantum Research Center (NVAQC) in Boston, aimed at integrating quantum computing with AI supercomputers, despite Huang's recent claims that practical quantum systems are still decades away. The NVAQC will collaborate with institutions like the Harvard Quantum Initiative and MIT, with aims to solve challenging problems in quantum computing and enable accelerated quantum supercomputing. Recommended read:
References :
Jaime Hampton@BigDATAwire
//
NVIDIA's GTC 2025 showcased significant advancements in AI, marked by the unveiling of the Blackwell Ultra GPU and the Vera Rubin roadmap extending through 2027. CEO Jensen Huang emphasized a 40x AI performance leap with the Blackwell platform compared to its predecessor, Hopper, highlighting its crucial role in inference workloads. The conference also introduced open-source ‘Dynamo’ software and advancements in humanoid robotics, demonstrating NVIDIA’s commitment to pushing AI boundaries.
The Blackwell platform is now in full production, meeting incredible customer demand, and the Vera Rubin roadmap details the next generation of superchips expected in 2026. Huang also touted new DGX systems, highlighting the push towards photonic switches to handle growing data demands efficiently. Blackwell Ultra will offer 288GB of memory. NVIDIA claims the GB300 chip brings 1.5x more AI performance than the NVIDIA GB200. These advancements aim to bolster AI reasoning capabilities and energy efficiency, positioning NVIDIA to maintain its dominance in AI infrastructure. Recommended read:
References :
Chris McKay@Maginative
//
NVIDIA's GTC 2025 event showcased significant advancements in AI infrastructure, highlighting the Blackwell Ultra and Rubin architectures, along with several related technologies and partnerships. Jensen Huang, Nvidia CEO, delivered a keynote address outlining the company’s vision for the AI-powered future, emphasizing improvements in processor performance, network design, and memory capabilities. The Blackwell Ultra GPUs are being integrated into DGX systems to meet the rising demands of AI workloads, especially in inference and reasoning.
NVIDIA is also expanding its offerings beyond chips with the introduction of desktop AI supercomputers for developers. The DGX Station, powered by the GB300 Blackwell Ultra Superchip, aims to bring data center-level AI capabilities to a compact form factor. Nvidia introduced Dynamo, an open-source inference software engineered to maximize token revenue generation for AI factories deploying reasoning AI models. The presentation emphasized a clear roadmap for data center computing, advancements in AI reasoning capabilities, and bold moves into robotics and autonomous vehicles. Recommended read:
References :
@tomshardware.com
//
Nvidia has unveiled its next-generation data center GPU, the Blackwell Ultra, at its GTC event in San Jose. Expanding on the Blackwell architecture, the Blackwell Ultra GPU will be integrated into the DGX GB300 and DGX B300 systems. The DGX GB300 system, designed with a rack-scale, liquid-cooled architecture, is powered by the Grace Blackwell Ultra Superchip, combining 36 NVIDIA Grace CPUs and 72 NVIDIA Blackwell Ultra GPUs. Nvidia officially revealed its Blackwell Ultra B300 data center GPU, which packs up to 288GB of HBM3e memory and offers 1.5X the compute potential of the existing B200 solution.
The Blackwell Ultra GPU promises a 70x speedup in AI inference and reasoning compared to the previous Hopper-based generation. This improvement is achieved through hardware and networking advancements in the DGX GB300 system. Blackwell Ultra is designed to meet the demand for test-time scaling inference with a 1.5X increase in the FP4 compute. Nvidia's CEO, Jensen Huang, suggests that the new Blackwell chips render the previous generation obsolete, emphasizing the significant leap forward in AI infrastructure. Recommended read:
References :
|
BenchmarksBlogsResearch Tools |