Huawei chairman claims Ascend AI chips surpass Nvidia sales in China
Huawei rotating chairman Eric Xu claims its Ascend neural processing units have exceeded Nvidia in Chinese market share.
Company
Nvidia is a technology company facing shifting dynamics in the global artificial intelligence market while actively expanding its software and hardware security offerings. In China, US export controls and local policy pressures have led to increased competition, with Huawei claiming its Ascend AI chips have surpassed Nvidia in Chinese market share. To address security concerns regarding autonomous AI agents escaping test environments, Nvidia has launched the Open Agent Safety Platform. Backed by major industry partners like Microsoft, Anthropic, and SpaceX, this platform combines OpenShell software for sandboxing with Sentry, a hardware-level data processing unit watchdog designed to isolate rogue agents within milliseconds. In addition to security, Nvidia continues to release new hardware and software tools. The company sells the RTX PRO 6000 Blackwell Workstation Edition GPU for 16,000 dollars, targeting professionals with 96 gigabytes of memory and 24,064 CUDA cores. On the software front, Nvidia has released Nemotron 3 Diarization, a free 100 million parameter model for real-time speaker identification, and developed the SoL-Pi system, which reduces coding agent token usage by up to 49 percent.
25 stories · Updated
Huawei rotating chairman Eric Xu claims its Ascend neural processing units have exceeded Nvidia in Chinese market share.
Nvidia released a new safety platform designed to contain rogue AI agents within milliseconds as security concerns mount.
Nvidia released the Open Agent Safety Platform on Monday, featuring an open-source sandbox and a hardware kill switch to stop autonomous AI agents.
Nvidia has introduced a new safety platform designed to quarantine rogue AI agents within milliseconds.
Nvidia has launched a new safety platform combining OpenShell software and a hardware watchdog called Sentry to secure AI agents.
Nvidia is releasing OpenShell and Sentry as part of a new open source platform designed to contain and monitor autonomous AI agents.
Thieves stole two trailers emblazoned with Nvidia and PlusAI logos in California, only to discover they were filled entirely with 20,000 pounds of sand each.
NVIDIA sells its powerful RTX PRO 6000 workstation GPU for $16,000 to target professionals instead of gamers.
Nvidia released Nemotron 3 Diarization, a free 100 million parameter AI model that identifies up to eight speakers in real time.
A new Nvidia paper details the SoL-Pi system, which automatically optimizes the control layer of coding agents to reduce token usage by nearly half.
Nvidia's upcoming free AI model might tie the UAE closer to American technology while testing local strategies for sovereign artificial intelligence development.
Google, Nvidia, and Anthropic have joined a new coalition to use Emerald AI software to secure grid space for new data centers.
Countries across the Global South are buying American hardware while adopting Chinese open-source AI models to avoid technological dependency.
Apple is reportedly planning a return to the server market with custom ARM processors and Nvidia technology, though a launch is not expected until 2029.
Cornelis Networks and Delos Data have launched new open networking alternatives to challenge Nvidia's dominant NVLink technology.
Taiwanese prosecutors have indicted nine people, including workers from NVIDIA and Super Micro, for smuggling restricted AI servers to China.
Nvidia researchers reveal that a custom software harness allows Claude Opus 5 to score 100 percent on the ARC-AGI-3 benchmark.
Nvidia is working with Goldman Sachs, BlackRock, and other financial giants on a 500 billion dollar financing effort.
Ukrainian military intelligence discovered an Nvidia Jetson Orin module inside the wreckage of a Russian S-71 Monochrome cruise missile.
Nvidia plans to invest $150 billion annually in Taiwan, solidifying its role as a key AI hub despite US efforts to bolster domestic production.
At GTC 2026, Jensen Huang said Nvidia expects $1 trillion in purchase orders through 2027 between Blackwell and Vera Rubin. He unveiled the full Vera Rubin platform, the Groq 3 LPU from its $20B acquisition, and teased the next-gen Feynman architecture for 2028.
Nvidia CEO Jensen Huang said the company investments in OpenAI and Anthropic are likely its last, citing upcoming IPOs. But the real story may be the growing rift between the two AI labs and the circular economics of these deals.
Nvidia is investing $2 billion each in Lumentum and Coherent to develop silicon photonics for next-generation AI data centers, signaling that moving data between chips is becoming the next big bottleneck.
A senior Trump administration official confirms DeepSeek used Nvidia Blackwell GPUs to train its next model, likely violating U.S. export controls.
Nvidia is close to finalizing a $30 billion stake in OpenAI as part of a funding round that could value the AI startup at $730 billion.