Latest News Analysis
OpenAI might be building its own chip, but it’ll still be dependent on Nvidia
Premium OpenAI is widely understood to be developing a custom AI accelerator with Broadcom, part of a $10 billion partnership to design chips specifically for OpenAI’s internal workloads.
Amid Intel's deals, Intel Foundry remains notably absent — 18A and 14A are on the way, but success isn't guaranteed
Premium It has been a busy few weeks for semiconductors, but among various announcements and rumors, Intel Foundry has remained conspicuously absent.
Data centers to account for 9% of electricity demand in the U.S by 2035
Premium Regulatory approval and standardization could take years to get them off the ground, though.
OpenAI's significant investments raise more questions than answers
Premium Long-term profitability is still a vague idea.
OpenAI makes flurry of deals in drive towards for-profit model
Premium OpenAI is making major moves to secure future GPUs and accelerate its hardware plans in a flurry of major deals worth billions of dollars, though questions remain about its future funding.
China bets on DUV as EUV blockade reshapes chipmaking — but it won't dethrone ASML's advanced lithography, for now
Premium With EUV locked out by sanctions, China is leaning on DUV multi-patterning and fledgling domestic scanners.
Why packaging is a huge part of Nvidia's $5B Intel deal — Foveros could speed up market delivery
Premium Foveros and EMIB are doing more than stacking dies; they’re accelerating Nvidia’s roadmap.
Exploring the RTX Pro 6000D, Nvidia's China-only GPU, which is now banned from sale
Premium Nvidia's RTX 6000D might be banned from sale in China, but what lies inside the GPU itself? We explore why it's not as good a deal as grey-market RTX 5090s in the region.
UK cosies up to big tech with $42 billion data center and AI investment deal
Premium UK inks deals with OpenAI, Microsoft, Nvidia, Google, Coreweave, ScaleAI, and more.
Nvidia tipped to be TSMC's first A16 customer, ahead of Apple — Feynman GPUs could make full use of GAA transistors and backside power
Premium Nvidia will be the first customer to use TSMC’s A16, a 1.6nm-class process that marries gate-all-around (GAA) transistors with backside power delivery.