To The Moon

Every earnings season has a dominant theme. This quarter, it wasn't AI demand. It was whether the companies spending hundreds of billions on AI infrastructure could continue doing so without breaking their financial model. After listening to the latest earnings calls from $Microsoft(MSFT)$ $Alphabet(GOOG)$ $Meta Platforms, Inc.(META)$ my takeaway is simple: The AI buildout isn't slowing. If anything, the conviction behind it is getting stronger. It starts with nearly $2 trillion of backlog. The combined commercial backlog across the major hyperscalers now stands at roughly $2 trillion. A significant portion of those contracts is expected to convert into revenue o

$SK hynix(SKHY)$ just reported what should have been a dream quarter. Revenue reached ₩79.3 trillion, up 257% year over year. Operating profit surged 557%, while operating margin climbed above 76%, marking the strongest profitability level in the company’s history. And yet, the stock dropped sharply. The reason is simple: The results were incredible. The expectations were even higher. A Record Quarter Wasn't Enough Anymore For most companies, numbers like these would trigger a major rally. But SK Hynix has become one of the biggest symbols of the AI infrastructure boom. Investors were already pricing in an extraordinary future, not just strong current results. Revenue came in slightly below elevated expectations, and operating profit also missed s

One of the biggest long-term challenges in semiconductor manufacturing has always been lithography equipment. Now, China's domestic ecosystem appears to have reached another important milestone. According to industry reports, a Shanghai-based semiconductor equipment company has begun manufacturing immersion DUV (Deep Ultraviolet) lithography systems, with the first batch expected to be delivered to local chipmakers for production verification. While the initial production volume remains small, the development represents another step forward in China's efforts to build a more complete semiconductor equipment supply chain. Capital Markets Reacted Immediately The report also triggered a sharp reaction across semiconductor equipment stocks. Following the news, shares of several major equipment

China's memory industry just had its biggest moment yet. On its market debut, $Cxmt Corporation(688825)$ surged 531.06%, closing the morning session at RMB 54.65 per share and reaching a market capitalization of approximately RMB 3.655 trillion. The rally pushed the company past $ICBC(01398)$ to become the most valuable stock in China's A-share market. It also briefly became worth more than $BABA-W(09988)$ , underscoring just how much enthusiasm investors now have for China's AI semiconductor sector. But this isn't just another blockbuster IPO. It's a reflection of how investors are pricing the future of China's AI and memory industry. Why Did the Market Reac

The AI infrastructure race appears to be entering another phase. According to reports, $NVIDIA(NVDA)$ is negotiating a $250 billion financial guarantee to support OpenAI's lease of SoftBank's planned 10-gigawatt AI data center campus in Ohio. If completed, the project could cost roughly $500 billion, making it the largest AI infrastructure project announced to date. More Than a Chip Supplier The proposed structure goes well beyond selling GPUs. Nvidia would reportedly provide financial guarantees that help lower financing costs for the project, while separately discussing financing for OpenAI's GPU purchases, which could total another $350 billion. In other words, Nvidia isn't simply supplying hardware—it is helping make the infrastructure itself

The latest earnings cycle delivered dramatically different outcomes across Big Tech. $Alphabet(GOOGL)$ dropped about 7%, wiping out roughly $300 billion in market value. $Tesla Motors(TSLA)$ fell 14.5%, erasing nearly $200 billion. $IBM(IBM)$ has lost more than 25% over the past two weeks. Then came $Intel(INTC)$. Instead of disappointing the market, Intel delivered one of its strongest quarters in years, beating expectations on revenue, earnings, margins, manufacturing progress, and guidance. After the report, the stock climbed in after-hours trading. But the biggest takeaway isn't simply that Intel had a good quarter.

$Taiwan Semiconductor Manufacturing(TSM)$ just delivered another record quarter, but the bigger story may be what comes next. The company said overseas manufacturing—particularly its expanding U.S. footprint—will weigh on margins over the next several years. At the same time, reports indicate TSMC plans to raise foundry prices by up to 10% beginning in 2027. Taken together, the message is straightforward: higher costs are increasingly likely to be shared with customers rather than absorbed entirely by TSMC. U.S. expansion comes with a price TSMC continues to invest heavily outside Taiwan, including a major expansion in the United States. Management acknowledged that as these new fabs ramp up, gross margins are expected to face modes

The AI boom is creating one of the largest investment cycles in technology history. But while investors focus on AI revenue growth, model performance and data center expansion, another number is becoming increasingly important: How much future financial commitment is being created behind the scenes? A recent analysis of financial filings from $Alphabet(GOOGL)$ $Microsoft(MSFT)$ $Amazon.com(AMZN)$ $Meta Platforms, Inc.(META)$ $Oracle(ORCL)$ suggests that AI-related obligations—including GPU contracts, data center leases and infrastructure agreements—have grown rapidly and now re

$Advanced Micro Devices(AMD)$ has unveiled Helios, its first rack-scale AI platform, marking one of the company's biggest strategic moves in the AI infrastructure market. The significance isn't simply another GPU launch. It's that AMD is moving beyond selling individual accelerators and is now offering a complete AI computing system that combines GPUs, CPUs, networking and software into a single rack-level solution. The Battle Has Moved Beyond GPUs For years, AI competition centered on who had the fastest accelerator. That is changing. Large cloud providers increasingly purchase complete AI systems rather than standalone chips, making system integration just as important as raw computing performance. Helios reflects this shift by combining 72 next-

Moonshot AI's Kimi K3 has attracted far more demand than expected. According to the company, user demand over the past 48 hours has pushed its GPU resources close to capacity. To maintain service quality for existing subscribers, Moonshot has temporarily paused new memberships while it adds more computing capacity. Existing subscribers are not affected, and the company plans to reopen subscriptions in stages. Moonshot also announced that future memberships will be split into separate plans for its general AI services and coding products, allowing compute resources to be allocated more efficiently. The Bottleneck Is No Longer AI Adoption The significance of this announcement extends beyond one product launch. Kimi K3's rapid adoption suggests that the challenge is no longer convincing users