AI hardware

Nvidia has added $150 billion to its stock buyback program, which is described as the largest ever undertaken by a company. The announcement underscores the chipmaker's enormous cash generation from demand for its artificial intelligence processors, returning capital to shareholders at an unprecedented scale. Market watchers see the move as a signal of Nvidia's confidence in its continued dominance of the AI hardware boom.

AMD has agreed to an $8.2 billion deal to bring in Fei-Fei Li, the Stanford AI researcher widely known as the 'Godmother of AI'. The move signals AMD's push to compete more aggressively with Nvidia in artificial intelligence chips and computing. Analysts are weighing what the deal means for the AI hardware race.

Nvidia is developing a watchdog chip designed to sit alongside every AI agent, monitoring its behaviour and preventing harmful or unauthorised actions. The idea reflects growing concern in the tech industry about autonomous AI systems acting without adequate oversight. Discussion centres on whether hardware-level safety mechanisms can keep pace with rapidly deployed agentic AI, and whether Nvidia's move will become an industry standard for AI safety.

A filing shows ByteDance, TikTok's Chinese parent company, gained access to more than 2,000 Nvidia B200 AI chips through Nscale's data center in Norway. A Singaporean subsidiary, Spring, contributed 73% of the UK neocloud provider Nscale's 2025 revenue. The arrangement highlights how US export restrictions on advanced chips to China can be circumvented through cloud capacity hosted in third countries, raising questions among observers about enforcement of AI hardware controls.

Nvidia has announced an Open Agent Safety Platform, a reference framework for continuous in-silicon monitoring of AI agents. The offering, detailed on the company's developer blog, is aimed at tracking agent behaviour in real time at the hardware level, reflecting growing industry concern over safely deploying autonomous AI systems.

New work shows microcontrollers, previously considered too limited for generative AI, can now run a diffusion model and a 289-million-parameter large language model on-device. Electronics and embedded systems outlets are covering the achievement, which points to generative AI moving beyond cloud servers and desktop GPUs onto cheap, low-power hardware.

Memory module maker Netlist has asked US authorities to ban imports of Micron chips used in AI computing systems built for Google and Nvidia, escalating a long-running patent dispute between the two companies. The request, if granted, could disrupt supply chains for high-performance memory components central to the AI boom. Industry watchers are tracking whether trade officials will intervene in a case with wide implications for the semiconductor sector.

Nvidia is working with insurance companies to address the financial and liability risks tied to artificial intelligence, according to a new report. The chipmaker, whose hardware underpins much of the AI boom, is said to be exploring coverage for potential harms linked to AI systems. Insurers are increasingly being drawn into debates over who bears responsibility when AI causes damage.

Memory module maker Netlist has asked US trade authorities to ban imports of Micron chips used in artificial intelligence computing systems built for Google and Nvidia, escalating a long-running patent dispute between the two companies. The move could disrupt supply chains for AI hardware if regulators side with Netlist, drawing attention across the semiconductor and cloud computing industries.

Baidu has spent roughly 15 years developing its own AI chips in-house, a long-running effort to reduce reliance on foreign semiconductor suppliers. The company now faces a public test as its chips are judged against market expectations, amid intense competition in China's AI hardware sector and ongoing US export restrictions shaping demand for domestic alternatives.

Nvidia is in discussions with insurance companies to develop loans backed by its AI chips, according to the Financial Times. The arrangement would let customers finance purchases of Nvidia's expensive AI hardware using the chips themselves as collateral. The move reflects the enormous cost of AI infrastructure and Nvidia's effort to keep demand for its chips growing among buyers facing financing constraints.

Chinese AI firm DeepSeek says it used open-source software tools running on Huawei Ascend chips, underscoring its reliance on domestically developed hardware and software amid US export restrictions on advanced chips. The claim is drawing attention to China's push for a self-sufficient AI stack built on homegrown silicon and open-source frameworks.

PaleBlueDot AI is seeking $600 million in private credit to finance purchases of computer chips, according to Bloomberg News. The move reflects growing demand for expensive AI hardware, with startups increasingly turning to debt markets rather than equity funding to secure the processors needed for training and running artificial intelligence models.

The EU Chips Joint Undertaking has opened two calls for projects focused on developing AI hardware. The initiative is part of European efforts to strengthen domestic semiconductor capabilities and support the hardware needed for artificial intelligence applications. Funding details and eligibility criteria are expected to draw interest from chipmakers, research institutions and technology companies across Europe.

Chinese AI firm DeepSeek has partnered with Huawei to develop chip programming tools, a move that reduces its dependence on Nvidia's hardware. The collaboration signals a deepening push by Chinese AI companies to build on domestic silicon amid US export restrictions on advanced chips. The alliance pairs one of China's leading AI labs with its top domestic chipmaker.

A new initiative is combining AI-driven chip design with pilot production lines for III-V compound semiconductors, aiming to speed the path from research to electronics manufacturing. The effort highlights growing investment in advanced semiconductor materials beyond silicon, as demand for AI hardware accelerates and governments and industry seek to strengthen domestic chipmaking capabilities.

A Seeking Alpha analysis examines Arista Networks (NYSE: ANET), arguing the company occupies a critical layer of networking infrastructure beneath the semiconductor and AI chip boom. The piece frames Arista's switches and software as essential plumbing for data centers powering artificial intelligence workloads, a position that continues to draw investor attention amid the AI-driven rally in tech hardware stocks.

UL Solutions has introduced a certification program for direct-to-chip liquid cooling equipment used in AI data centers. The certification aims to establish safety and performance standards for cooling hardware as AI workloads drive higher power densities that air cooling struggles to manage. The announcement was carried across industry and regional news outlets, reflecting growing demand for validated liquid cooling solutions.

Cloud provider General Compute is adding Cerebras chips alongside its existing Nvidia GPUs, aiming to speed up AI coding agents, Fortune reports. The move reflects growing demand for faster inference hardware as developers increasingly rely on autonomous agents that need rapid, repeated model responses. It also signals rising competition for Nvidia as alternative chipmakers like Cerebras find commercial footing.

Chinese AI startup DeepSeek has unveiled software tools built for Huawei AI chips, a move that could allow its models to run without Nvidia hardware. The development highlights China's push toward technological self-sufficiency amid US export controls on advanced chips, and signals growing momentum for Huawei's Ascend processors as domestic alternatives in the AI computing market.

A developer is sharing field notes from running Anthropic's Claude Code as an autonomous agent every day on a single Windows PC, highlighting performance numbers from the ongoing experiment. The write-up covers how the AI coding agent handles real programming tasks unattended. Readers are debating whether autonomous coding agents are reliable enough to be part of an everyday developer workflow.

Chinese memory maker CXMT has started mass producing its fifth-generation G5 DRAM chips, and PC makers are already placing orders. The move signals CXMT is closing the technology gap with established suppliers like Samsung, SK Hynix and Micron, and could affect global DRAM pricing and supply dynamics as demand for memory from PC and AI hardware keeps rising.

Chinese AI firm DeepSeek has open-sourced its full infrastructure stack built on Huawei's Ascend chips, claiming one-to-one parity with Nvidia's platform for AI training and inference. The release gives developers worldwide access to a complete, non-Nvidia hardware and software pipeline, underscoring China's push for self-sufficient AI compute amid US export restrictions. Commenters are weighing the parity claim against Nvidia's mature ecosystem.

Chinese AI firm DeepSeek has partnered with Huawei to develop chip programming tools, moving away from Nvidia's platform. The collaboration signals DeepSeek's effort to build its AI stack on domestic hardware, amid US export controls restricting Chinese access to advanced Nvidia chips. The move is being read as a further step in China's push for technological self-sufficiency in semiconductors and artificial intelligence.

Chinese AI firm DeepSeek has unveiled software designed to make Huawei chips capable of running AI models, positioning it as an alternative to Nvidia's hardware. The move highlights China's push to build a domestic AI computing stack less dependent on US technology, and analysts are watching closely for what it means for Nvidia's dominance in the AI chip market.

DeepSeek and Huawei are reportedly teaming up to develop AI computing technology that moves beyond Nvidia's hardware. The partnership would pair DeepSeek's AI models with Huawei's domestic chip ecosystem, a notable step for China's push toward technological self-sufficiency amid US export restrictions. Details about the scope and timeline of the collaboration remain limited.

A new wave of AI-ready computers is on the way, and Swedish tech outlet Computer Sweden reports they will likely cut ongoing cloud costs for running AI workloads, even though the machines themselves will not be cheap. The article urges IT buyers to weigh lower token expenses against higher upfront hardware prices when planning upcoming PC purchases.

Memory module maker Netlist has asked US authorities to ban imports of Micron chips used in AI computing systems built by Google and Nvidia, according to a Reuters report. The request escalates a long-running patent dispute between the two memory companies. An import ban could disrupt supply chains for advanced AI hardware, drawing attention across the semiconductor industry.

Commentary is asking where profits in the artificial intelligence boom actually land once semiconductor sales are done. The argument raised is that chipmakers capture early value, but the longer-term gains may go to software developers, cloud providers and companies that turn AI into services and products. The piece feeds a wider debate about whether today's heavy AI hardware spending will translate into durable returns across the industry.

Featherless, a serverless AI inference provider, is making the case that heavyweight infrastructure is overkill for small, routine AI workloads. The company uses the pizza-delivery analogy to argue that many applications can be served cheaply on demand rather than keeping large GPU capacity running constantly. The argument has drawn attention among developers weighing cloud costs for machine learning deployment.