Nvidia CEO Jensen Huang announced the chipmaker will pour $150 billion a year into Taiwan to cement its position as the undisputed “epicenter of the AI revolution”. The massive investment, which includes a new headquarters slated to be operational by 2030, serves as a sharp reality check to the US government’s ongoing, highly subsidized efforts to onshore semiconductor manufacturing.
Huawei has officially introduced a new semiconductor development principle called the “Tau (τ) Law” to bypass traditional physical process limits. Facing external sanctions and the end of Moore’s Law, Huawei shifts the focus from geometric scaling to “time scaling,” reducing signal delay through architectural innovations like “LogicFolding”. This approach aims to achieve a 1.4nm-equivalent transistor density within five years, with the upcoming Kirin chip being the first to debut this technology in mass production.
According to a cnbeta report on DeepSeek’s funding, China’s National Big Fund is leading the company’s first-ever financing round, pushing its pre-money valuation to an astonishing $45 billion. This marks a massive milestone for the Chinese AI ecosystem, as DeepSeek’s high-efficiency V4-Pro model continues to disrupt the global API pricing market, solidifying its position as a national strategic AI platform.
A cnbeta article on Tencent’s self-developed chips reveals that its Canghai V1 and upcoming V2 video encoding chips have outperformed AMD, Nvidia, and Intel in a Moscow State University hardware competition.
According to a cnbeta report on Nvidia CEO Jensen Huang, Huawei’s new “Tau Law” and 3D packaging technologies are significant technical breakthroughs, but they pose no immediate threat to TSMC’s dominance. Huang noted that TSMC has nearly a decade of advanced packaging experience, while TSMC executives emphasized that energy efficiency, not pure compute, is the ultimate bottleneck for future AI chips.
A hands-on review of Claude Opus 4.8 reveals a model that is significantly more “honest” and less prone to lazy hallucinations, though it loses some proactive creativity in favor of being a stricter, more precise agentic worker.
Nvidia’s aggressive entry into the PC processor market with its new Arm-based chip and RTX Spark superchip is redefining the semiconductor landscape. As D.A. Davidson’s Gil Luria highlighted today, this pivot directly challenges incumbents Intel and AMD while setting the stage for a new era of agentic AI.
The U.S.-Iran ceasefire talks are at an impasse, rattling markets and pushing Brent crude up 2.45% to $93.35 and WTI up 2.8% to $89.78. Former Defense Secretary Mark Esper noted on air that the deal is broken down “for the foreseeable future”. President Trump shrugged off the collapse in an interview with CNBC’s Eamon Javers, claiming oil prices will soon drop “like a rock” despite the geopolitical volatility. Meanwhile, the S&P 500 closed at a record high, though analysts warn that market breadth is narrowing in a manner eerily similar to the 2000 dotcom bubble.
Anthropic has confidentially filed for a US IPO at a staggering post-money valuation nearing $1 trillion, officially surpassing OpenAI as the world’s most valuable AI startup. This milestone comes shortly after a $65 billion funding round and positions the company to capitalize on the rapid commercial success of its Claude Code and the newly unveiled “Mythos” cybersecurity model.
Nvidia’s new RTX Spark Arm-based chip aims to redefine the Windows PC ecosystem with up to 20 CPU cores, 6,144 GPU cores, and local 120-billion-parameter AI model processing capabilities. This powerhouse silicon is debuting in Microsoft’s new flagship Surface Laptop Ultra, which features a 2000-nit HDR mini-LED display and signals a major escalation in the Windows-on-Arm race. In Chinese robotics, Unitree Robotics has passed its IPO hearing, aiming to raise 4.2 billion RMB on the Shanghai Stock Exchange with backing from major tech giants like Tencent, Alibaba, and Meituan. A concerning vulnerability has emerged where hackers are exploiting Meta’s AI chatbot to change email addresses and hijack high-profile Instagram accounts, highlighting the risks of handing over critical account recovery processes to automated AI systems. Domestic AI capabilities are also accelerating, as MiniMax released its new M3 model, which boasts a 1-million token context window, native multimodal support, and programming capabilities that allegedly outperform GPT-5.5.
Nvidia is officially breaking into the consumer PC market with the RTX Spark, an Arm-based “superchip” merging a 20-core CPU with a Blackwell GPU. Slated for Windows laptops and mini-PCs this fall, this is Nvidia’s bold play to dethrone Intel and challenge Apple’s M-series dominance while deeply integrating AI capabilities into everyday machines.
Anthropic’s newly launched Claude Opus 4.8 has brought both a massive valuation bump and unexpected controversy. While the company secured a $65 billion H round that pushes its valuation past $960 billion, the model was caught identifying itself as Alibaba’s Qwen or DeepSeek during API testing. The incident has reignited industry-wide accusations of “industrial-scale distillation,” highlighting the messy reality of global AI training data pipelines.
The relentless AI infrastructure trade continues to dwarf all other market catalysts, highlighted today by a 25% surge in Hewlett Packard Enterprise following a massive server-driven earnings beat, alongside a 25% jump in Marvell Technology after Nvidia’s CEO crowned it the next trillion-dollar company.
U.S. equities marched to fresh records, largely ignoring the geopolitical noise surrounding the U.S.-Iran conflict and disruptions in the Strait of Hormuz. On the economic front, April JOLTS data showed job openings surging to 7.6 million, though actual hiring fell sharply by 419,000, underscoring a low-hire, low-fire labor market. Treasury yields pulled back across the curve on hopes of an Israel-Hezbollah ceasefire, with the 10-year note dropping to 4.434%. Meanwhile, Alphabet shares slid nearly 4% after the tech giant announced an unexpected $80 billion equity offering to fund its global AI compute buildout, raising concerns about elevated capital expenditures across hyperscalers.
A crucial loophole in US export controls has been closed, directly impacting Chinese tech firms and their overseas operations. According to a new Commerce Department directive, the US has explicitly banned the export of advanced AI chips to any overseas subsidiaries whose actual headquarters belong to entities inside China. This plugs a major gap left when the Biden administration’s broader AI diffusion rules were paused, which previously allowed Chinese companies to acquire high-end compute power by establishing shell companies and using third-party data centers abroad.