Semiconductors

If you’ve been waiting for the AI hardware story to stop being “Nvidia, Nvidia, and also Nvidia,” circle April 14, 2026. That’s when Meta and Broadcom went public with an expanded partnership to co-design multiple generations of Meta’s MTIA accelerators through 2029, anchored by a 1-gigawatt initial deployment and a path to multiple gigawatts after that. The headline spec: the first AI silicon built on TSMC’s 2nm process. This isn’t a routine vendor press release. It’s the loudest signal yet that the era of hyperscalers handing blank checks to Jensen is ending. ...

There’s a factory rising in the snow-covered plains of Hokkaido, Japan, and the government just bet another $4 billion that it can change the future of AI. On April 11, 2026, Japan’s Ministry of Economy, Trade and Industry (METI) approved ¥631.5 billion ($4 billion) in fresh subsidies for Rapidus — a semiconductor startup that most of the industry has politely called “ambitious” and privately called “impossible.” The new infusion brings total government backing to a staggering ¥2.6 trillion ($16.3 billion), making Rapidus one of the most heavily state-funded chip ventures in history. ...

Nvidia’s deepest moat was never the silicon. It was CUDA — the software ecosystem that made every AI developer on Earth, including China’s, completely dependent on Nvidia’s way of doing things. You could build a faster chip, but if developers had to rewrite their entire codebase to use it? Dead on arrival. Huawei just found the side door. The Ascend 950PR, paired with Huawei’s overhauled CANN Next software stack, has reportedly won over ByteDance and Alibaba — two of China’s largest AI consumers. After years of Beijing practically begging its tech giants to go domestic, Huawei may have finally built a chip they actually want to use. ...

Everyone’s fighting over GPUs. Alibaba just changed the question. On Tuesday, Alibaba’s DAMO Academy unveiled the XuanTie C950 — a 5-nanometer server processor built on open-source RISC-V architecture. It’s the highest-performing RISC-V CPU ever made. But the interesting part isn’t the benchmarks. It’s the thesis behind the chip: that AI agents need fundamentally different silicon than AI chatbots. While Nvidia, AMD, and Intel wage war over who can build the biggest parallel processor for training models, Alibaba is making a deliberate bet on what comes after training. And the logic is harder to dismiss than you’d think. ...

After 35 years as the Switzerland of semiconductors — licensing chip designs to anyone with a checkbook — Arm Holdings just crossed the Rubicon. It built its own chip. Not a demo. Not a reference design you’ll never see in production. A 136-core data center processor called the AGI CPU, fabricated on TSMC’s 3nm process, with Meta signed on as the debut customer. This isn’t incremental. This is tectonic. The Hardware: 136 Cores of Pure Intent The specs read like Arm had something to prove. ...

The lights shooting into the Austin sky on Saturday night weren’t aliens. They were Elon Musk doing what Elon Musk does best — staging an event so audacious that you can’t look away, even if you’re not sure you believe a word of it. Inside the defunct Seaholm Power Plant in downtown Austin on March 21, Musk officially launched Terafab — a joint venture between Tesla, SpaceX, and xAI to build what he calls “the most epic chip-building exercise in history by far.” The price tag: an estimated $20–25 billion. The goal: producing one terawatt of computing power per year, with 80% of it destined for space. ...

Seven words on X. That’s all it took to detonate the semiconductor world. On March 14, Elon Musk posted: “Terafab Project launches in 7 days.” Over 866,000 views within hours. And if Tesla actually pulls this off, it could be the most ambitious vertical integration play in the history of chip manufacturing. What Terafab Actually Is Think “Gigafactory, but for chips” — except Musk says it’s “like giga but way bigger.” Confirmed during Tesla’s January earnings call, Terafab would combine logic processing, memory storage, and advanced packaging under one roof. Full vertical integration at a scale that currently only exists at TSMC and Samsung. ...

The AI chip wars just got a lot more interesting. Broadcom CEO Hock Tan dropped a number so staggering it made Wall Street’s collective jaw hit the floor: AI chip revenue exceeding $100 billion in 2027. Not total revenue — just the AI chip slice. Last quarter, that number was $8.4 billion. This isn’t just a big number. It’s a signal that the way we build AI infrastructure is fundamentally changing — and Nvidia might not be the only kingmaker anymore. ...

Tomorrow evening, Nvidia drops the most consequential earnings report of 2026. And it’s not just about one chipmaker’s quarterly numbers — it’s the entire AI investment thesis going on trial. Nvidia isn’t a semiconductor company anymore. It’s the barometer for the AI economy. The single stock that tells you whether hundreds of billions in AI infrastructure spending are paying off or spiraling into speculative excess. With NVDA sitting at $191, a $4.62 trillion market cap, and an AI market rattled by Anthropic-triggered selloffs, the pressure is immense. ...

What if instead of running an AI model on a chip, you turned the model into the chip? That’s the bet Taalas just went public with — and the numbers are making the entire semiconductor industry sit up straight. This 25-person startup out of Toronto emerged from stealth with $169 million in funding and a working product called the HC1: a chip that hard-wires a large language model directly into silicon transistors. No software stack. No HBM memory. No liquid cooling. Just raw, physics-level inference. ...