AI chips

Anthropic is building an in-house team to design AI chips for its Claude model. The company will hire engineers to co-design chips and models to improve performance and scalability.

Blackstone is proposing a second debt deal to finance Anthropic's AI chips, with an initial package of $36 billion, surpassing the previous $35 billion deal from two months ago.

Taiwan's GDP is growing rapidly due to increased U.S. imports of AI chips, but experts caution potential challenges from Trump and China could emerge.

Elon Musk stated during Tesla's earnings call that building Optimus, a humanoid robot, will be the hardest product to manufacture at scale in Tesla's history due to its complexity, lack of existing supply chains, and challenges in dexterity, reliability, and AI integration. Tesla must develop components in-house and solve technical hurdles to achieve mass production.

Tesla's significant investments in AI, robotics, and self-driving technology are reducing operating profits despite record vehicle deliveries. The company plans to spend over $25 billion on capital investments this year and borrow up to $30 billion for projects like robotaxis and Optimus robots, with CEO Elon Musk expressing confidence in long-term returns.

Trump's push for American-made AI chips has led TSMC to invest $200 billion in U.S. manufacturing, impacting the company's margins. The investment was announced following Trump's return to power in 2025.

Nvidia’s H200 chips have resumed limited imports into China, while China’s chip exports nearly doubled in the first half of the year, primarily consisting of mature logic integrated circuits for consumer electronics and automotive sectors. The US leads in advanced AI chips, whereas China is becoming a dominant player in mass-market legacy chips.

Chinese semiconductor design firm Biren Technology has introduced next-generation 'supernode' solutions using optical data transmission to connect thousands of AI chips in a single cluster. The move highlights the competitive race among AI infrastructure companies to scale computing power as AI models advance.

Nvidia CEO Jensen Huang mediates internal competition for limited AI chips as demand from data centers grows. The company's automotive division, a strategic priority, competes for GPU resources alongside other teams, with Huang occasionally resolving disputes. Resource allocation balances immediate business needs and long-term goals like autonomous driving.

ASML has increased its sales forecast for the second time this year due to strong demand for AI chips from its customers.

The article discusses how powerful trends like AI chips and Bitcoin can lead to severe market corrections despite their strong growth, highlighting the tension between paradigm shifts and speculative bubbles.

ZML, a French AI startup endorsed by Turing Award winner Yann LeCun, has released ZML/LLMD, a free software product designed to speed up AI inference across multiple AI chips and reduce costs.

South Korea's AI chip industry, driven by Samsung Electronics and SK Hynix, is generating significant wealth, but concerns persist about unequal profit distribution. A legal dispute over the valuation of SK Hynix shares could alter business tycoon Chey Tae-won's asset value by billions.

Amazon is designing its own AI chips for Echo and Fire TV, with Hardware chief Panos Panay stating the company builds end-to-end silicon for critical devices. The article hints at a new portable AI gadget coming soon.

Amazon is developing custom AI chips for devices like Echo and Fire TV, as well as future products, according to hardware chief Panos Panay. The company aims to experiment with AI-powered gadgets through these specialized chips.

The 2026 COMPUTEX event in Taipei, Taiwan, featured Intel CEO Lip-Bu Tan and NVIDIA CEO Jensen Huang showcasing next-generation AI chips and robotics. The trade show highlighted Taiwan's role as a leading hub for semiconductor and AI supply chain innovation.

South Korea is investing $880 billion in AI chips and data centers. The move highlights the country's focus on advancing artificial intelligence infrastructure.

The EU has approved the Commission to sign the US-led 'Pax Silica' initiative, which targets China in the context of AI chips. The decision followed months of internal negotiations within the Council.

The article highlights two stocks that could benefit from the next generation of AI chips. The Investing Club hosts its 'Morning Meeting' daily at 10:20 a.m. ET.

Asia investors are seeking alternative ways to invest in SpaceX after being excluded from its upcoming IPO. SpaceX has secured a $920 million monthly cloud computing deal with Google, which will utilize 110,000 Nvidia GPUs to support Google's AI models.

Berlin and Warsaw are collaborating on AI chips. The article also mentions Ireland's tech priorities, Meta's lobbying efforts, and the development of AI gigafactories.

China is planning a $295 billion AI data center buildout, requiring state firms to source at least 80% of technology, including AI chips, from local suppliers. This initiative aims to exclude foreign companies like Nvidia and AMD.

Google is ordering millions of AI chips from Intel as TSMC struggles to meet demand. This has led major tech companies to seek alternative manufacturers.

Business Insider's Olivia Nemec visited Intel's advanced chip factory in Oregon, where strict contamination controls prevent human skin, hair, and dust from damaging semiconductors. The facility, which produces critical chips for AI and everyday technology, enforces rules like avoiding makeup and wearing $1,000 bunny suits. Semiconductor sales are projected to reach $1 trillion by 2027.

The U.S. government is working to increase domestic production of printed circuit boards to reduce reliance on China, as security concerns arise over Chinese-made circuit boards used in AI chips.

The EU Commission's Technological Sovereignty Package aims to enhance European AI chip production and cloud services, focusing on defense applications. The initiative is led by the EU executive to strengthen technological capabilities.

Senators Elizabeth Warren and Andy Kim criticized the Trump administration for potentially allowing advanced AI chips to be exported to overseas units of Chinese firms. They requested Commerce Secretary Howard Lutnick to testify to Congress, following guidance from the Department of Commerce to close an export loophole.

The Trump administration is closing a loophole in U.S. export restrictions requiring licenses for selling advanced AI chips to firms with Chinese parent companies, even if those firms are not located in China or other restricted countries. The Commerce Department issued new guidance clarifying this requirement.

High-bandwidth memory is a critical component for AI chips, providing the necessary data flow. Demand for this memory has significantly boosted Micron's market value, briefly pushing it over $1 trillion.

Samsung workers in South Korea approved a government-backed bonus deal to avoid a strike, with over 60% of the workforce eligible for $370,000 bonuses amid high demand for AI chips.

Melius Research has upgraded its outlook for Dell and Apple, while Amazon's custom AI chips are gaining competitive traction in the market. Jim Cramer highlights these developments as key factors to monitor in Tuesday's stock market activity.

AMD CEO Lisa Su met with Chinese Vice-Premier He Lifeng, signaling potential relaxation of US chip export restrictions to China. The meeting suggests possible reopening of high-end semiconductor and AI chip sales to select Chinese firms. The discussion focused on welcoming foreign investment and positive outcomes from recent US-China summits.

CNBC's Jim Cramer believes Nvidia should be allowed to sell AI chips in China, as it will keep Chinese companies reliant on American technology. This decision can help Nvidia's stock thrive. Cramer's opinion suggests a positive outlook for Nvidia's business in China.

Meta has signed a multi-year agreement to use Amazon's AI chips in a deal valued at billions of dollars, according to AWS. The partnership highlights growing collaboration between major tech firms in AI infrastructure.

Tesla is planning to build AI chips using Intel's unfinished 14A process for its Terafab project, as revealed by Elon Musk during a recent earnings call. The move highlights Tesla's need to develop its own silicon despite relying on an unproven manufacturing node.

Google Cloud has launched two new AI chips (TPUs) that are faster and cheaper than previous versions. However, Google continues to use Nvidia's technology in its cloud services for now.

Cerebras, an AI chip startup, has filed for an IPO after securing a deal with Amazon Web Services to deploy its chips in data centers and a reported $10 billion agreement with OpenAI.

Federal prosecutors have charged six individuals for smuggling billions of dollars' worth of AI chips to China, exposing loopholes in U.S. export controls. Cases involve Super Micro Computer and Thai business contacts, highlighting pervasive smuggling networks in Southeast Asia. The U.S. and China are both reshaping AI chip supply chains for national security, but enforcement gaps persist.

Anthropic disclosed a $30bn run rate and plans to use 3.5GW of Google's new AI chips. Broadcom is building these chips for Google, which Anthropic intends to leverage for its operations.

Nvidia reported an 85% revenue increase, demonstrating strong financial performance driven by demand for AI chips. However, the company maintained its exclusion of China from its business outlook, continuing to navigate geopolitical restrictions on semiconductor exports.