Orbit Post Sitemap

AI money is circulating more and more like a loop The AI industry chain is forming a complex relationship: Chip manufacturers provide investment or financing support to model companies and data centers, which then use these funds to purchase chips and computing power, with orders ultimately returning to the chip manufacturers' accounts. NVIDIA is at the center of this network. It not only sells GPUs but also invests in model companies and collaborates with Wall Street institutions to provide funding for customers building AI data centers. *AMD also has arrangements with major customers that include procurement agreements, equity incentives, and strategic cooperation coexisting Supporters believe this is not "left hand to right hand" but a normal financing method for new infrastructure construction. AI data centers require huge investments and have long payback periods; chip manufacturers connecting capital, equipment, and customers can accelerate market expansion. However, when suppliers simultaneously become shareholders and financiers of customers, it becomes harder to judge whether revenue comes from genuine end-user payments or internal industry chain fund circulation. If AI applications fail to generate cash flow for a long time, risks may transmit simultaneously through equity, debt, and chip orders. The key to judging whether this round of investment is healthy is not how many GPUs are sold, but whether money continues to flow in from outside the loop: whether enterprises renew subscriptions, whether AI can increase profits, and whether data centers can generate stable cash flow. End-user demand paying the bill is called industry synergy; relying solely on chip manufacturers to keep injecting capital looks more like circular financing. #英伟达深入AI资本链，协同与风险如何平衡 #Consumption momentum weakens, September policy still constrained by inflation Recently, US data has become increasingly interesting: inflation is falling, consumption is also cooling down, but the market finds it harder to predict the Fed's next move. Retail sales in July fell by 0.6% month-on-month, and consumer confidence was below expectations. Looking at this data alone, I would lean further towards maintaining the interest rate in September rather than continuing to raise it. The logic is actually very simple: weakening consumption means demand is cooling down, combined with the previous simultaneous easing of CPI and PPI, the Fed now has less strong necessity to keep stepping on the economic brakes. One-year inflation expectations have slightly rebounded. If residents start to re-form the expectation that "prices will continue to rise," the Fed's biggest concern about inflation stickiness has not been truly resolved. So going forward, compared to retail sales, I will pay more attention to inflation data and statements from Fed officials. The same applies to $BTC. The most comfortable scenario now is not that the US economy suddenly worsens, but that the economy slowly cools + inflation continues to fall + the Fed stops raising rates. This is the truly favorable soft landing environment for risk assets. I currently still lean towards BTC having one more downward process to find support, but if inflation continues to cool and the September rate hike expectations further fade, I will also reassess this judgment Tether has finally completed the "Big Four" audit, is the trust crisis of $USDT over? After years of market skepticism, Tether has finally delivered a more substantial response. KPMG U.S. has completed the first full independent audit of Tether International's 2025 financial statements and issued an unqualified opinion. This is different from previous reserve attestations. This audit covers the balance sheet, income statement, cash flow, asset ownership, and valuation. As of the end of 2025, Tether's reserve assets exceed liabilities by about $6.8 billion. In response to external doubts, CEO Paolo Ardoino directly replied: "Honestly, I don’t care." His confidence comes from $USDT having undergone a real stress test. During the severe market turmoil in 2022, Tether handled about $7 billion in redemptions within approximately 48 hours without stopping payments. But transparency controversies are not over yet. Currently, the full financial statements and KPMG audit report have not been made public. Tether believes that as a private company, it is not obligated to disclose all data like a publicly listed company. Now the market will no longer question: Does Tether dare to accept a full Big Four audit? The real trust in stablecoins is how the market reacts when situations arise, whether this 1:1 dollar can truly be redeemed. #Tether首次完整审计：透明度成焦点 #Consumption momentum weakens, September policy still constrained by inflation The worse retail is, the more BTC should rise Last night, July retail sales month-on-month were -0.6%, expected +0.1%, turning negative for the first time in nine months. Excluding autos -0.3%, excluding autos and gasoline still -0.2%, overall cooling. Tax refunds are spent, Prime Day was moved to June and overspent, American households are running out of cash Inflation cooperates here, Wednesday CPI 3.4%, core 2.5%, PPI dropped from 5.5% to 4.7%, all below expectations. Economy falters, inflation goes down, if you were Powell, would you raise rates? CME shows September rate hike probability dropped from 50% to 30%, a pause in September is certain, the discussion now is whether to hint at rate cuts by year-end On the market, $BTC is grinding near 63000 above the dense trading zone at 65000, a volume breakout above can push short stop-losses far; support at 62000 below. $ETH is stuck at 1880, failed to hold 1900 but with strong elasticity, breaking 1900 targets 2000, support at 1850. $XAU is the strongest, London gold spot at 4375, New York gold futures at 4432, surged to 4454 last night, repeatedly testing highs, 4400 support not broken means bullish The last BTC doubling was partly due to spot ETFs, but the real trigger was rate cut expectations. Now it’s the same flavor. Of course, if a real recession hits, BTC will fall too, but that’s the second half. In the expectation phase, bad news is good news Before the September FOMC, every bad data release is candy for risk assets Fear and Greed Index at 29, fear. This number has been hovering in the 20-40 range for almost two months, neither extremely fearful nor returning to neutral. A dull knife cutting meat is the most tormenting. It won't make you panic sell, but it will slowly wear down your patience until one day you can't take it anymore and liquidate everything with one click. In 2019, when BTC dropped from 14,000 to 6,000, the fear index hovered around 30 for three months, then suddenly surged. The bottoming phase is less about judgment and more about who gives in first. $SKHYNIX Last night the storage sector was full of drama, SanDisk alone hit the daily limit up, while Micron and Hynix just played dead😂 To be clear, no matter how explosive the profits are, if there’s no buyback or dividend, it’s just messing with shareholders. Investors aren’t stupid; money flows to the sensible kids. To wait for Hynix to have a chance, unless Temasek really steps in and negotiates governance with the South Korean government, only then would it be worth betting on follow-up news. Let’s talk after the pullback is in place, this script is still unfolding🤔 Institutions collectively bullish! NAND shifts from a cyclical product to a core infrastructure theme, $SNDK breaks above 1649, rebounding 60% in two weeks—can it still be chased? Brothers, SanDisk has truly been redefined by institutions this round—JPMorgan resumed coverage with an overweight rating, target price 2250; Citi maintains a buy rating, target 2100; Morgan Stanley, though cautious, also acknowledges AI demand supporting earnings. The core logic is that large model inference drives demand for KV Cache and enterprise-grade SSDs, turning NAND from a consumer electronics cyclical product into an AI infrastructure mainstay. The market votes with money: up another 6.5% on Friday, a cumulative 35% rebound this week, and over 60% gain in just over two weeks. Price surged from 1300 to 1650, a very clear bullish trend. Looking at the 1-hour candlestick chart, price is hugging the upper Bollinger Band at 1665, MACD shows a high-level convergence with signs of a death cross, RSI values range between 61-73. Short-term overbought but no divergence yet; trend intact but risk of chasing higher is accumulating. Key levels: Resistance: 1665-1680 Support: 1600-1620 Objective view: The collective institutional re-rating signals a mid-term logic shift, not short-term speculation. Profit-taking may occur near 1650, and a pullback is an opportunity. Trading strategy: For existing positions, scale out gradually around 1660-1680 to lock in profits. For those without positions, enter long if the pullback holds the 1600-1620 range. If volume breaks below 1600, stay on the sidelines. The NAND story is being rewritten, but good entry points come with patience. #闪迪投资者日后股价大涨，长期目标待验证 #交易之声：你的经验值得被听到 🔥 $BTC vs $ETH — Institutional Capital Is Starting to Tell a Different Story One trend I’m watching closely right now is the divergence in ETF flows. Bitcoin spot ETFs saw strong demand earlier in August, with roughly $850M of net inflows during the first week. Since then, flows have become more volatile. Ethereum ETFs, meanwhile, have continued attracting relatively steady attention. That doesn’t mean institutions are abandoning $BTC. The bigger story is capital allocation becoming more selective. Bitcoin has been the clear institutional gateway into crypto for years. But Ethereum is increasingly becoming part of the allocation conversation as its ecosystem, on-chain activity, and institutional use cases continue to develop. The signal isn’t one week of inflows or outflows. It’s whether the divergence persists. If $ETH continues attracting capital while $BTC ETF flows remain unstable, we could be entering a phase where institutions are becoming more selective about where they gain crypto exposure. So the next question isn’t simply: “How high can BTC go?” It’s: “Where will institutional capital put the next dollar?” 👀 That shift in capital allocation could matter far more than short-term price movements. #WeakConsumptionFedSplit #OpenAIAnthropicRace #SKHynixCapexSurge $BTC $ETH Short-term bullish, but volatility remains. Since August, Tesla's stock price has strengthened, mainly due to cooperation with SpaceX, energy and AI deployments, and Robotaxi advancement; however, the shelving of the economy model, high capital expenditures, and regulatory pressure will limit upside potential. Stock price and August performance - Recent performance: Stock price strengthened after entering August, rising 0.58% on August 11 and 0.68% on August 12, with market sentiment leaning optimistic - Volatility background: In late July, stock price was pressured due to expanded capital expenditures, negative free cash flow, and regulatory investigations - Valuation drivers: Market expectations for future businesses such as autonomous driving, AI, and robotics are key to supporting valuation Cooperation with SpaceX - Roadster and "flight capability": The new Roadster is jointly developed by Tesla and SpaceX, planned to showcase a limited edition with "flight" capability at the Texas test site; demonstration is uncertain but topical - "Terafab" chip factory: Co-building an advanced chip factory in Texas with a first-phase investment of $16.8 billion; SpaceX builds its own gas power generation and large battery arrays to ensure power supply, aiming to support computing power for Optimus and FSD - Solar and energy manufacturing: Both parties are simultaneously advancing solar capacity expansion, targeting 100 GW annual manufacturing capacity within three years, and evaluating factory expansion in the U.S. - Analyst merger expectations: Some Wall Street analysts believe the two companies may merge in the future, possibly advancing in 2027 via an all-stock deal Energy and battery technology - V2L function expansion: Expanding vehicle-to-load functionality for Model Y Premium in the U.S. to enhance usage flexibility - Battery recycling progress: Key battery material recycling rate in U.S. operations has exceeded 90%, piloting more efficient recycling technologies to strengthen closed-loop and cost advantages - Energy storage product iteration: The third-generation large energy storage product Megapack 3 is planned for mass production in 2026, consolidating leadership in the energy business AI and robotics - Accelerated Optimus mass production: Dedicated Optimus production line installed at Fremont factory, with an annual capacity target of 1 million units; the third-generation humanoid robot is expected to be released and produced in the second half of 2026 - AI large model integration: In the Chinese market, OTA access to ByteDance's "Doubao" large model enhances in-car voice and interaction capabilities - Robotaxi and Starlink: Robotaxi model Cybercab integrates Starlink V5 and 5G combined connectivity to ensure continuous communication in weak or no network environments Regulatory and operational updates - Nevada testing permit: Authorized by Nevada to test up to 10 Robotaxis in specific areas but has not yet obtained commercial operation permits - Regulatory investigations and recalls: The U.S. National Highway Traffic Safety Administration has launched suspension failure investigations on nearly 1.2 million Model 3/Y vehicles; over 20,000 Model 3/Y vehicles recalled due to low beam brightness issues Challenges and risks - Economy model shelving: Entry-level economy model (Model 2/"Redwood") project was halted by Musk in April 2025, with resources redirected to AI and Robotaxi, making it difficult to boost sales through low prices in the short term - Capital expenditure and profit pressure: Capital expenditure expected to exceed $25 billion in 2026 for new factories and AI infrastructure; free cash flow turned negative in Q2 2026, putting short-term profit and cash flow under pressure Investment and trading recommendations - Focus on catalysts: Closely track Roadster flight demonstration, "Terafab" factory progress, Optimus mass production pace, and Robotaxi commercialization - Risk monitoring: Continuously assess capital expenditure and free cash flow, regulatory investigation progress, and the impact of Model 2 shelving on sales and profit margins - Positioning and strategy: Maintain phased accumulation and stop-loss amid optimistic expectations to avoid chasing highs; consider repricing opportunities brought by technology and cooperation progress during pullbacks $TSLA #消费动能转弱，9月政策仍受通胀制约 Current end-consumer demand recovery is slowing, with weak consumer willingness and domestic demand repair falling short of previous market optimism. The market originally expected a round of easing policies in September, but inflation stickiness has become the biggest constraint, locking the space for significant easing, making it difficult for policies to fully stimulate the economy. 1. Signals of weakening consumption momentum 1. Discretionary consumption remains weak; major consumption categories like automobiles and home appliances show weak recovery, daily consumption is mainly essential, with insufficient willingness for excess spending. 2. Residents tend to save and are cautious about future income expectations; even small consumption subsidies have limited pull effect. 3. The service sector's impulse rally has faded; cultural and tourism consumption declined after summer, lacking new consumption growth points. 2. How inflation constrains September easing Inflation has not fully declined to a range that allows worry-free easing; some price components remain resilient: • Large interest rate cuts and increased stimulus could trigger price rebounds, bringing new inflationary pressure. • Therefore, monetary policy will be cautious, with a lower probability of large rate cuts; more use of fiscal tools and targeted support rather than broad easing. Simply put: although domestic demand is weak, prices have not yet given policy the "green light" to loosen fully. 3. Major asset scenario analysis 1. Stock market: expectations for a broad bull market cool down, shifting to structural trends. Pro-cyclical consumer sectors are unlikely to see widespread rallies; funds will continue to cluster around AI and high-growth sectors. 2. Bond market: easing expectations are lowered, yield decline space narrows, the one-sided bull market phase ends, shifting to oscillation. 3. Commodities: weak domestic demand suppresses downside space, but inflation floor prevents sharp drops, resulting in overall range-bound oscillation. 4. Two key indicators to watch going forward 1. CPI inflation data for August-September; if inflation falls, policy easing space will reopen. 2. Types of policies implemented in September: if mainly special bonds and industrial subsidies, it indicates targeted easing; if rate cuts and reserve requirement ratio cuts appear, it means constraints have been lifted. The 635-day streak ends, XRP falls below $1 — Whales step in to buy ETF, but the “digital gold” narrative is unraveling --- 📊 1. Real-time Price Overview: 635-day streak ends, $1 support broken As of August 15, XRP was priced at $1.0052 on the Investing.com OKX platform, down 0.45% in 24 hours, with an intraday range of $0.9872 to $1.0101. · Market Cap: approximately $63 billion · 24-hour Trading Volume: approximately $1.1 billion · 52-week Range: $0.9872 to $3.1853 · Year-to-date: down 45.44% · 1-year decline: 67.34% · 7-day decline: 3.31% · 1-month decline: 9.71% The key signal: XRP has fallen below the psychological $1 mark, hitting a low of $0.9915, ending the 635-day streak of holding above $1. The current price is near the 52-week low. 📉 2. Market Review: From “iron bottom” to “breakdown,” the 635-day defense line collapses in a day XRP had held above $1 since November 2024. But on August 11, this 635-day record was broken. Since then, XRP has fluctuated around $1 — briefly recovering above $1.01 on August 13, but falling below again on August 15, hitting a low of $0.9872. The price remains below all major moving averages. 🔥 3. Market Drivers: Structural contradictions are tearing apart XRP’s narrative 1. XRPL adoption hits new highs, but XRP is being “bypassed” (the core variable) XRPL network adoption continues to reach new highs, with on-chain RWA total value approaching $4.06 billion, a net increase of about $2.5 billion in six months. Aviva Investors (managing $351 billion in assets) has received approval from the Central Bank of Ireland to launch tokenized funds on XRPL. The problem: Ripple’s cooperation with its top ten institutional clients in 2026 all choose to settle in RLUSD, not XRP. XRPL is growing, but XRP cannot benefit. Institutions value the infrastructure itself, not the token asset. 2. ETF inflows plunge 93%, whales forced to “step in” In the first half of August, XRP spot ETFs attracted only $3.27 million in net inflows, a sharp drop of about 88% from July’s $27.29 million. On August 10, 11, and 12, there were zero net inflows for three consecutive days. ETF total assets under management have fallen below the $1 billion mark, down to $942 million. Meanwhile, whales increased their XRP holdings by 72 million in the past 24 hours, accumulating over 452 million in recent weeks. Whales are using real capital to absorb liquidity withdrawn from ETFs. 3. CLARITY Act delayed, regulatory uncertainty extended The CLARITY Act failed to pass before the Senate recess. If passed, XRP could be officially classified as a “digital commodity,” potentially rising 50%-100% to $1.50-$1.55. The delay means prolonged regulatory uncertainty. 4. Technicals: Weekly RSI enters the second historical oversold zone XRP’s weekly RSI has entered the second-ever historical oversold zone — the last time this happened, XRP recorded a gain of about 1,085%. The monthly RSI has dropped to 40, below the previous major low range of 44-47.5. 📈 4. Technicals and Key Levels Investing.com’s comprehensive technical rating is a “strong sell.” XRP continues a bearish trend with lower highs and lower lows. Key Supports: · $0.93: first critical zone · $0.87: correction endpoint noted by multiple analysts · $0.75-$0.66: further downside support · $0.70-$0.90: main support range · $0.62: extreme target highlighted by Ali Martinez Key Resistances: · $1.03: must reclaim and hold to improve short-term structure · $1.10: must hold above to make rebound more convincing · $1.48: breakout would trigger $727 million short squeeze Liquidation Risk: If XRP falls near $0.89, approximately $157 million in long positions could be liquidated. 💎 5. Summary XRP is at a critical moment: the end of a 635-day streak, loss of $1 support, and a 45% year-to-date decline. Bullish factors: weekly RSI’s second historical oversold (last time led to 1085% gain), whales continuously accumulating (72 million in one day, 452 million over weeks), XRPL network adoption hitting new highs (RWA at $4.06 billion), some analysts target $3-$5 or even $14. Bearish factors: 635-day streak broken, Ripple’s top ten clients all switching to RLUSD settlement (XRP “bypassed”), ETF inflows down 93%, CLARITY Act delayed, technicals show “strong sell.” $1 is the last psychological defense line — holding it still leaves bulls a chance to turn things around; a confirmed break below opens downside to $0.87-$0.93, even $0.70-$0.75. XRP stands at a crossroads for its next direction. $XRP #闪迪投资者日后股价大涨，长期目标待验证 Let them pull, let them pull, the breeze brushes the hillside. The day the chips loosen is the real time to act. After the Sandisk investor day, the stock price rose 13.7% in two days and did not fall back the next day. The Korean stock market was even stronger, rebounding 22% in ten days, with Samsung and SK Hynix leading the semiconductor sector upward. The same AI hardware logic, but the gains on both sides are not on the same level. The difference is simple: Korean stocks are a position replenishment after a deep oversell—leverage has been wiped out, chips cleared, so the rebound is naturally fast. Sandisk’s rise is a fundamental revaluation, supported by solid numbers like AI storage demand, a $14 billion buyback, and long-term gross margin targets. One is filling a gap, the other is lifting valuation; their nature is different, and so is their sustainability. But both share the same hidden concern: supply is expanding, can demand keep up? Sandisk’s FY2028 mid-to-high double-digit growth and 80% gross margin, SK Hynix’s new NAND line won’t start production until the second half of 2026. All expectations are front-loaded; performance fulfillment is still on the way. The ten-day rebound has swallowed all previous declines; what remains is either story potential or performance potential? Fast gains don’t necessarily go far. Among the 22% rebound in Korean stocks, those entering are already thinking about when to exit. The chips at the bottom are always changing hands; some come in, some go out. I haven’t moved my position. It’s not that I haven’t seen the rise, I’m waiting for a better entry point. Both Sandisk and Korean stocks are telling the AI story, but truly good companies are those that can keep rising after they have already risen. Let them pull, let them pull, the breeze brushes the hillside. The day the chips loosen is the real time to act. $SNDK $MU $BTC #海力士扩产提速，资本开支能否兑现回报 #OpenAI与Anthropic估值竞赛升温 Prophetic prediction! Brothers, look at this trend, exactly the same as what I said yesterday! $APR was smashed through, dropping directly from 0.63 to 0.21, nearly a 50% drop in 24 hours. I really admire myself, shorted in at 0.459, ran at 0.460, because right after shorting it instantly surged to 0.47, resulting in an instant loss of -130%—at that time I felt no profit and couldn’t hold on, so I immediately lowered my position and ran. Looking back now, I really want to slap myself. This precise timing but perfectly missing the main downtrend wave is truly a classic move in the crypto world. To be honest, I said yesterday that this surge was contract funds igniting the market, with open interest hitting $25.45 million, net inflow over $4.8 million, a typical low-cost hype tactic. Small-cap coin, new story, a few million can triple the price. But if no one takes over, it can only return the same way. Now RSI has dropped from 99.6 straight down to the oversold zone, EMA5, EMA10, EMA20 all pressing overhead. After falling to 0.278, volume picked up significantly, indicating someone is trying to catch the falling knife. But at this position, who dares to say this is the bottom? The 90-day gain is still 98%, the 180-day gain is still 179%, profit-taking hasn’t cleared out yet. I ran from my short yesterday out of fear, but the big picture was right. A coin propped up by sentiment, where is the sustained buying support? Surge, sideways, distribution, crash—the script hasn’t changed a bit. My current view: short term might consolidate and catch a breath, but until the trend reverses, don’t rush to catch the bottom. Brothers who want to catch it, wait until it stabilizes above EMA20. Reaching out now is most likely catching a knife.$BTC suddenly receives dovish support! Wall Street is pulling back on rate hike bets. This time, it’s not the market guessing blindly, but U.S. consumer data clearly cooling down. U.S. retail sales in July fell 0.6% month-over-month, far below the expected 0.1% growth, marking the first decline in 9 months and the largest drop in 14 months. Core retail sales also dropped 0.4%. With consumption cooling, combined with weakening employment and inflation data, market concerns about the Fed continuing to raise rates naturally start to ease. The most obvious change is: Traders are pulling back on rate hike bets. The implied tightening for September has dropped from about 19 basis points at the end of July to about 9 basis points. The probability of a rate hike in September has also fallen from nearly 50% a month ago to about 30%. Simply put: A month ago, the market was worried about the Fed continuing to raise rates; now, a rate hike increasingly looks like a low-probability event. This is at least a somewhat positive signal for BTC, ETH, and the entire risk asset market. Because the retreat of rate hike expectations means one of the biggest pressures on liquidity is easing. But don’t rush to call a bull market just yet. While cooling consumption helps lower rate hike expectations, if the economy cools too quickly, the market will start to worry about a recession. So what really matters next is not just a single retail sales report, but whether: Inflation can continue to decline, the economy can remain resilient, and the Fed will gradually shift toward easing. If subsequent data continue to confirm this direction, the macro environment’s pressure on BTC will lessen. And once the market switches from "worrying about rate hikes" to "pricing in easing early," capital tends to rush in without prior notice. So what’s most worth watching now is not whether Wall Street is bullish or bearish today. But a core change: Rate hike expectations are retreating, and the pressure over risk assets is loosening. #消费动能转弱，9月政策仍受通胀制约 #OpenAI与Anthropic估值竞赛升温 #海力士扩产提速，资本开支能否兑现回报 $ETH $SNDK 最近，$SNDK 接连暴涨，有许多人都在蠢蠢欲动，想要做空。我在它初期上涨的时候就关注了，但是我一直没有写关于它的文章。为什么？因为我之前一直没有做空的把握，因为我找不到什么做空的依据。但是，我现在有一定把握了，因为我找到了一些依据。我认为，现在是可以做空的。 —————————————————— 我们看一下它的合约数据。我们先看它的合约多空比。对于它的合约多空比，我们需要关注三个时间节点。第一个时间节点是7月24日，当时它的合约多空比第一次触及低点。第二个时间节点是7月31日，当时它的合约多空比第二次触及低点。第三个时间节点是8月5日，当时它的合约多空比第三次触及低点。我们结合当时的K线去看，可以发现这三次触及低点都是当时$SNDK 价格的高点。而现在，它再一次触及低点了。但是，这并不能意味着闪迪现在就一定会跌了，它可能会触及更低点。我记得$OKB 最近的合约多空比就是在不断创下新低，所以我上面所说的只能作为一个参考。我做空也不主要是根据上面的数据。我们再看一下它的合约持仓量。可以发现，它目前的合约持仓量已经超过了7月30日的高位。这说明，现在确确 Retail consumption suddenly stalls, but US stocks are still hitting new highs US July retail sales data is out, showing a month-on-month decline of 0.6%, while the market originally expected a 0.1% increase, missing by 0.7 percentage points. This is the largest single-month drop since May last year, indicating that consumer spending is finally starting to falter. Consumption accounts for 70% of US GDP, so this data directly impacts the Q3 GDP growth rate. Many institutions are already reassessing whether to revise down their economic forecasts. Interestingly, the US stock market is completely unfazed. The S&P 500 index continues to hit record highs, breaking through 7800 points intraday for the first time and closing at 7799 points. The cooling PPI has pushed the probability of a September rate hike down to about 35%, which has actually eased the stock market. The trio of CPI, PPI, and retail data all point in one direction: inflation is retreating, consumption is retreating, and the necessity for rate hikes is also retreating. However, the US stock market and the crypto market are moving to completely different rhythms. The US stock market has entered a "bad news is good news" phase — a weakening economy means the Fed is less likely to raise rates, which supports valuations. Meanwhile, crypto is still bottoming out, with funds flowing into US stocks; in the existing market, BTC can only wait for liquidity to gradually return. Nokia rose nearly 15% this week, driven clearly by AI data center optical interconnect demand. Its Q2 optical network business revenue increased by more than 50%, making it one of the more stable performers among tech stocks in this wave. The cooling retail data should be positive for the medium to long term, but in the short term, BTC is still waiting for buying interest to return. #消费动能转弱，9月政策仍受通胀制约 ——$BTC #标普收盘再创新高，8000点预期升温 ——$NOK The biggest positive factor in August is Trump's midterm election. Many people have always believed that Bitcoin will definitely have a big surge during the August midterm elections because only an increase can better gain the support of Wall Street capital and capitalists for Trump, and only by raising cryptocurrency prices can more support be obtained. But the past historical data shows: in every round of midterm elections, BTC experiences a big drop!!! And the smallest drop occurred in 2022, with BTC's largest drop being 37.2%. When you see this real data laid out here, how do you feel? The midterm election's positive outlook is just what everyone hopes for in their hearts; it is not a genuine positive! Shouldn't you prepare your bullets, adjust your recent positions, and avoid risks? If, by chance, there is a drop of about 20%, how would you respond? Historically, every round from August to October has basically been a decline. I wonder if history will be rewritten in 2026? #CLARITY表决待定，SEC规则未落地 #特朗普因TruthSocial付费数据流遭起诉 #加密估值转向收入，BTC如何定价？ AI Capital Frenzy vs Crypto Bleeding: Are BTC and ETH Being "Drained"? Let's first look at two sets of numbers. In the first half of 2026, Microsoft, Amazon, Google, and Meta raised their full-year AI capital expenditure guidance to $725 billion, a 77% surge from $410 billion in 2025, burning through $130 billion in just one quarter. On the other hand, the US spot Bitcoin ETFs saw a net outflow of $4.5 billion in June, marking the worst monthly record since their 2024 launch, with BlackRock's IBIT alone losing $3.55 billion. Money is indeed moving, and the directions are clearly opposite. But to say AI "sucked away" money from crypto is only half true. The investors buying Nvidia stock and those buying BTC largely overlap—they are the same group chasing high beta returns. As the Nasdaq keeps hitting new highs fueled by AI narratives, and SpaceX's IPO draws tens of billions from retail investors in a day, marginal funds naturally flow first to stories backed by earnings and cash flow certainty. Crypto has no earnings to talk about, only narratives; once the narrative fades, redemptions become mechanical. The $4.5 billion outflow in June is less about AI draining funds and more about institutions cutting the "least cash-flow defensible" part of their portfolios amid shrinking risk appetite. Meanwhile, the Fed's dot plot under Kevin Warsh shifted and rate cut expectations were wiped out—that was the real main switch—because in a high interest rate environment, AI stocks have earnings to back them, crypto does not. So this is not a structural permanent diversion but a cyclical risk preference reshuffling. The evidence: funds haven't left the crypto market; they are just repositioning within it—June saw nearly $60 million net inflow into XRP ETFs, HYPE ETFs attracted a record $161 million, and on August 10, SOL ETFs recorded the largest single-day inflow since May. The money hasn't left, it's just become more selective. Which is more resilient, BTC or ETH? As of the morning of August 15, $BTC was around $63,100, down 2.8% for the week, with the August low defense line at $62,560 being repeatedly tested; if broken, the psychological $60,000 level is next. $ETH was at $1,878; on August 13, the day BTC ETFs saw a $131 million outflow, ETH ETFs actually had a net inflow of $6.72 million. This detail is worth pondering. BTC's current problem is its holding structure—profit-taking positions dropped to 51.4%, a three-year low, indicating many coins are underwater, so rebounds face selling pressure. ETH has fallen deeper, with more thorough leverage clearing, and staking yields provide long-term holders a reason to "hold and not sell," making its selling pressure more elastic. The Fear & Greed Index is 35, in the fear zone; SOL is at $75.3, DOGE at $0.069—these high beta assets have fallen harder, precisely indicating the market is undergoing defensive contraction. During contraction, the asset with the more stable nominal buy-side holds up better. In the short term, ETH's ETF fund flows look better than BTC's. The core contradiction is this: AI is drawing away "narrative premium," while the Fed is withdrawing "liquidity itself." For BTC to win back funds, it won't be by competing with Nvidia for attention, but by waiting for the moment interest rate expectations turn—that moment will likely see the first inflows return to BTC, which has the best liquidity and the most mature institutional channels; ETH will follow. Being drained is temporary, but the ranking has already changed.🛑 U.S. Consumer Suddenly "Hits the Brakes" The latest data shows that retail sales in July fell by 0.6% month-over-month, while the market expected a 0.1% increase. This is the largest drop since May last year. Even excluding autos and gasoline, sales still declined by 0.2%, indicating that the weakening consumption is not caused by a single category. At the same time, the University of Michigan's consumer sentiment index for August dropped from 55.2 to 51, the first decline in three months. The one-year inflation expectation rose to 4.3%. Real average hourly wages fell by 0.2% year-over-year — prices are rising, wages are not keeping up, and purchasing power is being squeezed from both ends. Consumption is the lifeblood of the U.S. economy, accounting for two-thirds of economic activity. With this data released, market expectations for multiple Fed rate hikes this year have clearly diminished. The two-year U.S. Treasury yield briefly fell below 4.10%. But the problem is not that simple. Consumption has dropped, but inflation has not. The one-year inflation expectation at 4.3% remains far above the 2% target. Oil prices remain high, with no relief in sight as the Strait of Hormuz remains closed and the Mandeb Strait is again on fire, keeping energy cost pressures elevated. Harmak just said "rate hikes are necessary now," inflation is widespread, and employment data will not change her focus. The Fed's internal hawks have not backed down at all. The weakening consumption momentum has indeed lowered the probability of a rate hike in September from a high level — CME data shows the probability of holding rates steady in September has risen to nearly 62%. But inflation stickiness is equally real. The Fed is caught in a dilemma between weak employment and stubborn inflation; which side prevails will only be clear in September. #消费动能转弱，9月政策仍受通胀制约 Weekend Position Handling and Next Week Strategy Switch｜LieLiu Intraday Trading｜Real Review + Trading Plan｜Professional Judgment BTC dipped with a wick last night, the lowest point did not hit my original stop loss at 61,988, so the long position is still held. But avoiding the stop loss doesn’t mean you can keep being greedy. 【Current Position】 This long position has been pushed to breakeven, with the price around 63,400—63,500. I will take profit directly and no longer gamble on weekend continuation. 【Weekend Plan】 No new intraday positions will be opened during the weekend. If the price later retraces to 62,000—62,500 and shows a confirmed stop in the decline, I will consider lightly entering a long position with a stop loss near 61,500 and a target of 64,000, which is the current upper boundary of the descending channel. No confirmation, no early entry. 【Next Week Strategy】 The daily chart is still forming a converging wedge, and the 4-hour chart is within a descending channel. My projection remains: August is treated as consolidation, next week may first test 64,000—64,500 upwards to attract bulls back, then choose to go down. So starting next week, I will gradually switch strategies: Shorting becomes the main direction, long positions reduced to half the usual size, only participating in rebounds; if 64,000—64,500 shows a spike followed by a drop, volume stagnation, or weakening of small-scale structure, I will start positioning trend short orders. Only if it further breaks below 61,500 will the downtrend acceleration be confirmed. Then in September, I will first watch 54,000, then 48,000, with interim rebounds being opportunities to continue seeking trend short positions. $BTC $ETH #消费动能转弱，9月政策仍受通胀制约【Invalidation Conditions】 If BTC breaks and holds above 64,500 with continued recovery near 65,150, the path of a bull trap turning to a drop will be delayed, and the short plan will be temporarily canceled. I am not asserting a September crash now, but writing the script in advance, waiting for price confirmation. Lock in profits over the weekend first. The market can wait, positions cannot be gambled.闪迪五天暴力暴涨35%，行情背后到底是什么驱动？是阴谋还是牛来了… 短短五个交易日，闪迪走出一波非常极端的上涨行情，累计涨幅达到35%，盘面大起大落，频繁插针来回扫止损，不管做多还是做空的交易者都被反复洗盘，短期波动已经脱离普通的震荡节奏。这一轮行情，有行业基本面消息加持，也有合约市场资金博弈放大波动。从消息层面来看，存储行业整体迎来利好催化，AI服务器对闪存存储的需求持续提升，行业大厂表态存储芯片或将迎来紧缺周期，带动整个存储板块情绪走高。同时闪迪公布和多家大客户签订长期锁量供货协议，未来两年很大一部分产能已经提前锁定，市场看好企业后续营收稳定性，给市场注入了乐观预期。但要注意，基本面只是导火索，短期35%的暴涨，很大一部分来自合约市场的逼空效应。前期积累大量空单，价格拉升过程中空单不断被清算，空单止损平仓又进一步推着价格往上冲，形成越涨越逼空的循环，把涨幅进一步放大。这就造成盘面一个很典型的现象：上涨阶段势头凶猛，但是一旦清算潮结束，买盘接力跟不上，就会出现快速回落，上蹿下跳，洗盘特征十分明显。现在市场分歧已经拉到最大，一部分人认为行业利好会继续推动行情走高，另一部分人认为 Is it possible to short SanDisk now? Direct conclusion: It is not recommended to short SNDK (SanDisk) naked at the moment, as it is a high-risk gamble and not a conventional reliable short opportunity (current price about $1641, closed +7.39% rebound on August 14). ✅ Bearish logic (reasons supporting shorting): 1. It is an extremely high beta cyclical stock, previously surged from 42.82 to 2354, with heavy sentiment premium. Once NAND spot price growth slows and cloud vendors' capital expenditures fall short of expectations, there is huge downside potential. 2. After a short-term rebound, it is still quite far from the 52-week high of 2354; NAND consumer demand is weak, price increase momentum is marginally weakening, only AI data center long-term contracts support the fundamentals. 3. Extremely volatile holdings and very high turnover rate; after positive news is realized, valuation can be quickly cut. ❌ The three biggest fatal risks of shorting now (important!): 1. Institutions are unanimously bullish with no sell ratings; many major banks have target prices above current price (average 2100+, highest up to 3000), and there could be upward guidance, large buybacks, or new cloud vendor long-term contract news at any time, causing a violent short squeeze. 2. Just had a continuous rebound (high gains in the past week), short-term trend is strong; shorting against the trend = counter-trend trading, which can easily lead to sustained short-term floating losses; storage sector often reacts to Micron and Samsung news, sudden positive news causing +10%+ in a single day is common. 3. Naked shorting with margin has unlimited loss: if it continues to surge above 2000+, losses will keep expanding; borrowing fees and broker recall risks also exist, which is very mismatched for your preference (if you don't want to take too much risk). 1. Technicals are weakening again: volume increased while breaking recent key supports (1520, 1400), rebound is weak. 2. Catalysts: NAND contract prices stop rising or turn down, AI storage order guidance downgraded, cloud vendors reduce capital expenditures. 3. Position: very small position, strictly limit total capital proportion, absolutely no leveraged heavy positions. 📌 Summary in one sentence: - If you want to be steady and avoid big losses: do not short now, the bullish narrative is still intact, counter-trend shorting has very low error tolerance. Today, when going out, everyone will probably check their phones and then sigh; the drop in BTC is a bit beyond expectations. The core reason is still the US retail data, with a glaring -0.6% figure, which is miles away from the expected 0.1%. The cooling of the US economy is obvious, with new employment turning negative, labor force shrinking, and now even consumption is lagging. On the BTC side, it couldn't break through 65,500, then reversed and fell below the 63,200 support. ETFs have had continuous outflows, with $329 million running out this week, indicating big money is rebalancing. Many are asking where this drop will bottom out; I personally think to watch 62,200 first. But if no one steps in to buy, even 60,000 is uncertain. Actually, I'm not worried that BTC will completely collapse, because in the $60,000 range, the willingness to buy is actually stronger than to sell, which can be seen in data monitoring. Many investors are actually waiting for a clear signal, such as progress in the US election or easing of US-Iran tensions. The current situation is: negative news is being digested, but new momentum hasn't emerged yet. As long as there is no catastrophic negative news, BTC is very likely to continue oscillating within a certain range. The current bottom can easily turn into a trap for those caught off guard.#SKHYNIX Expansion Accelerates, Can Capital Expenditure Deliver Returns SK Hynix $SKHYNIX is truly betting on the future of AI this time. Capital expenditure in the first half of the year increased by over 70% year-on-year, with continuous expansion in HBM, advanced packaging, and NAND. On the surface, this is to capture AI server demand, but in reality, it’s also a bet on a bigger question: can AI demand sustain this round of frenzied expansion? Of course, it’s still too early to simply say Hynix is expanding too fast. HBM is the most scarce link in the AI computing power industry chain; whoever can secure capacity and technology ahead has the chance to reap the dividends of AI demand growth in the coming years. But the risks are also very clear. Once AI demand cools down, the first to be hit won’t be storage prices, but capacity utilization. Because if capacity is built up in advance but demand doesn’t keep pace, idle capacity increases, fixed costs are diluted, which then impacts profit margins, and finally forces storage prices to be cut. So Hynix is not simply expanding now; it’s racing against AI demand. Expanding too slowly might miss orders; expanding too quickly might turn today’s high profits into tomorrow’s capacity pressure. What I’m focused on is not whether they dare to expand, but whether the new capacity in the coming quarters can truly be absorbed. As long as utilization and profit margins hold up, this is seizing an opportunity. Once inventory starts to pile up and utilization declines, that could be a signal of overexpansion. The above is just my personal opinion and does not constitute any investment advice! $AVNT I think the recent trend is more worth watching than simply "rising 17%." In early August, it was still around 0.078. After V2 launched, volume surged to around 0.11, and today's pullback still holds above 0.10. I have already established a long position. But looking at the market, I actually don't recommend chasing right now. The reason is simple: the first wave of sentiment brought by V2 has already been realized, and the next step is to verify real demand. After V2 launched, Avantis added a large number of RWA trading assets and further increased its OI capacity, significantly expanding market trading volume. This is also the core reason why I am bullish on AVNT. Not because of the three words "V2." Rather, I believe the RWA + Perp + Base route is starting to have the opportunity to transform from a concept into a real trading scenario. The market is now quite clear: 0.10 is the short-term lifeline for bulls. Holding this point, I believe there is still a chance to continue challenging 0.12. If 0.12 breaks out with increased volume, that would be a true structural breakthrough. But if 0.10 falls, especially if it continues to fall below 0.09, I will directly reassess this long position rather than holding it out with fundamentals. Currently, AVNT's OI is around $20 million, indicating that funds have come in, but there hasn't been any particularly extreme leveraged crowding yet. So my approach is simple: above 0.10, I'm on the bullish side. Breaking through 0.12, I think it accelerates. If 0.09 falls below it, I cut my losses. Additionally, there has been ongoing token activity recently 美联储9月利率决议的悬念正在消散——至少从衍生品市场的定价来看是这样。 CME“美联储观察”最新数据显示，美联储9月维持利率不变的概率为67.5%，加息25个基点的概率降至32.5%。这一分布表明市场已形成“9月大概率不动”的清晰共识。更远期的预期：10月不确定性上升将视线拉长至10月，市场的定价则呈现出更多的分歧：维持利率不变：53.3% 累计加息25个基点：39.8% 累计加息50个基点：6.8% 10月的“维持不变”概率较9月明显下降（67.5%→53.3%），说明市场预期美联储可能在10月采取行动，但方向仍不确定——加息可能性合计约46.6%（25bp+50bp），与按兵不动的概率接近持平。近期的数据如何影响预期？本周密集公布的宏观经济数据，是驱动加息预期从50%以上回落至32.5%的核心因素： CPI温和回落：通胀降温趋势确认 PPI全面低于预期：生产端压力缓解零售销售月率-0.6%：消费端意外疲软三组数据共同削弱了“美联储被迫加息”的叙事，为“按兵不动”的定价提供了数据支撑。对加密市场意味着什么？利率维持现状的概率优势明确，为金融市场提供了稳定的政策预期锚 Discussion about the US stock market approaching 8,000 points is heating up, but BTC is still weakening around $63,052, with ETH and SOL under pressure simultaneously. My judgment is that the current risk appetite has not broadly spread to the crypto market; funds remain concentrated on higher-certainty main themes like AI infrastructure and chip capital expenditures. Consumer weakness coexists with Fed divergences, and macro liquidity is temporarily insufficient to support broad valuation expansion. In the short term, crypto assets are more likely to remain volatile; a true signal of strengthening is not a single-day rebound but a sustained inflow of funds relative to the US stock market. Just my read, not advice.A Tale of Two Extremes: A Weekly Market Summary of A-shares and U.S. Stocks The biggest commonality between A-shares and U.S. stocks this week, in my opinion, can be summed up in two words: divergence. After a prior adjustment, A-shares entered a phase of volatile recovery. The Shanghai Composite showed relative strength, but the ChiNext Index and previously hot sectors like AI hardware, optical modules, and PCBs were clearly under pressure. Meanwhile, low-position sectors such as pharmaceuticals, military industry, non-ferrous metals, and consumer goods took turns becoming active. This indicates that capital has not truly exited the market but is undergoing a clear rotation between high and low sectors and rebalancing positions. The U.S. stock market continued to fluctuate near historical highs. The indices still appear strong, but internal divergence has begun to emerge. Especially in the AI sector, the market has shifted from the past mentality of "anything related to AI will rise" to focusing on revenue, profits, and cash flow. Palantir’s strong performance after exceeding earnings expectations is actually a good indicator: The next phase of AI competition is not about the story but about who can truly turn AI into profit. For A-shares, the AI supply chain including optical modules going overseas faces an additional layer of geopolitical risk. In the future, besides looking at orders and earnings, one must also consider overseas policies, supply chain restrictions, and the impact of financial pricing power. Therefore, I believe the market is not yet at a point where excessive pessimism is warranted. Trading volume remains, capital remains, but the old main themes are cooling down while new main themes have not yet fully formed consensus. At this stage, the most important thing is not to go all-in betting on a rebound but to control position sizes and observe where the capital is moving. Every intense turnover is a process of market repricing. Only when the old overcrowding subsides can new opportunities truly emerge SanDisk surges, you might be curious why short covering can push the price upward? The reason is simple: short sellers covering their losses is essentially a buy order. Those who short can't bear the losses and are forced to buy back, and the influx of buy orders further lifts the price, creating a positive feedback loop. Currently, SNDK has three layers of buying power: fundamental institutional capital, trend quantitative capital, and short covering. But it's important to distinguish that this is not an extreme short squeeze like GME; some short sellers have already exited. Short covering is just an accelerator for the rally, while the core driver of this round's rise is the better-than-expected earnings guidance. Trading plan: 1. Wait for a pullback: a drop to around 1550, or a consolidation between 1480-1520, stabilizing and showing a volume-increasing bullish candle, then consider light buying. 2. Breakout follow-up: you can also participate if it breaks above 1650 with volume, but be sure to control your position size. The first target is around 1730. #消费动能转弱，9月政策仍受通胀制约 $SNDK SK Hynix's single-quarter profit margin has surged to a historic high, with capital expenditures simultaneously rising to nearly 50 trillion KRW, yet the secondary market experienced a double-digit single-day plunge following the impressive earnings report. The stock price has sharply retraced over 20% from its peak, the spot market has quickly cooled down, and funds are choosing to concentrate on avoiding future heavy asset depreciation risks at the profit-taking point. Management has tied the large cash flow from the first half of the year with plans for new factories worth hundreds of trillions of KRW, attempting to lock in capacity absorption in advance through multi-year supply agreements signed with several core customers. When massive capital expenditures meet macro uncertainty, investors' risk appetite contracts, and position adjustments directly outweigh the performance security brought by long-term locked orders. If the multi-year orders from leading major customers are executed on schedule and delivery gaps persist, the suppressed valuation will see position inflows after the new capacity is smoothly launched. If terminal capital expenditures show marginal slowdown, the high capital expenditures will directly convert into depreciation costs on the income statement, lowering gross profit and triggering a new round of position reductions. As long as long-term orders experience widespread delays or defaults, the logic of locked-in long-cycle agreements will be completely falsified. The most important variable to watch in the coming week is the confirmation attitude of $SKHY's core customers at the supply chain end regarding additional prepayments and delivery rhythm. #CLARITY表决待定，SEC规则未落地 #英伟达深入AI资本链，协同与风险如何平衡 After seven weeks of sideways trading, Bitcoin breaks below $63,000 — ETF net outflows for three consecutive days, $63,500 “iron bottom” breached --- 📊 1. Real-time Price Overview: $63,000 level lost As of August 15, Bitcoin (BTC) fell below the key $63,000 level. Data from Investing.com Bitfinex platform shows BTC at $63,108, down 0.43% in 24 hours, with an intraday range of $62,530 to $63,504. Early morning quotes in Vietnam briefly dropped to $62,971–$62,982. · Market Cap: approximately $1.26 trillion · 24-hour Trading Volume: approximately $19.97 billion · 52-week Range: $57,877 to $126,110 · Year-to-date: down 28.03% · 7-day decline: 2.84% · 1-year decline: 46.28% Bitcoin has been trading sideways in the $62,000–$65,000 range for seven consecutive weeks, and today it finally broke below the lower boundary of this range. 🔥 2. Triple Drivers of the Decline: ETF Outflows, Whale Selling, Demand Weakening Driver One: Spot ETF net outflows for three consecutive days On August 14, the total net outflow from U.S. Bitcoin spot ETFs was $57.63 million, marking the third consecutive day of net outflows. In the past 24 hours, the lowest price touched $62,487. ARK 21Shares Bitcoin ETF led outflows with $58.8 million, followed by Fidelity FBTC with $55.1 million, totaling $113.9 million, accounting for 64.3% of total outflows. BlackRock IBIT saw $5.7 million outflows. The previous five trading days’ cumulative net inflow of $865 million has been offset by $332 million net outflows over nearly four trading days, accounting for 38%. Driver Two: Continuous selling by Jump Crypto and Bitdeer Jump Crypto has deposited about 1,560 BTC to Binance this week, worth approximately $99.2 million, and currently still holds about 1,410 BTC. Bitdeer produced and sold 263.4 BTC this week, maintaining zero Bitcoin holdings. Strategy (MSTR) sold about 1,690 BTC for stock buybacks. These three major selling pressures are resonating. Driver Three: Structural demand is weakening A 10xResearch report points out that Bitcoin is currently below the 7-day and 30-day moving averages, with a weekly change of -3.2%, indicating weakening structural demand rather than a single catalyst. Since August 9, open interest in perpetual contracts has been rising continuously, suddenly surging to 524,000 BTC on August 14, the highest level in nearly three months — futures leverage is accumulating while spot demand is retreating. 📉 3. Technical Analysis: $63,500 “Iron Bottom” Broken Current position: BTC has fallen below the key support at $63,500, with an intraday low of $62,530. The 50-day SMA dense zone between $64,000–$64,500 forms strong resistance. Dow Theory indicates the short-term uptrend was broken on August 10, with the higher low structure disrupted. Key Resistance: $63,500–$64,000 (previous support turned resistance) → $64,500 (50-day SMA) → $65,000 (psychological level) Key Support: $62,500–$62,800 (today’s low range) → $62,000 (psychological level) → $60,000–$61,000 (recent low range) CryptoQuant warns that historically, current losses have not yet reached the extreme levels observed before a true market bottom, so it is premature to declare a bottom. VanEck’s GEO framework (global liquidity neutral, on-chain activity neutral, ecosystem leverage constructive) supports the assessment that a bottom is forming, but the magnitude has not reached historical extremes. 💎 4. Summary The $63,500 “iron bottom” was breached today. ETF net outflows for three consecutive days, continuous selling by Jump Crypto and Bitdeer, futures leverage accumulation while spot demand retreats — these three forces pushed Bitcoin below the lower boundary of the seven-week sideways range. The direction is chosen: down. The next focus is whether $62,500 can hold. If it does not hold, $60,000–$61,000 will be the next key battleground. For bulls, reclaiming $63,500 is the first signal that the “alarm is lifted.” $BTC $SNDK surged 20% in two days, reaching a dense chip area above. The core contradiction lies in the short-term premium quickly pricing in the tight supply expectations before 2027. High-position chips are very prone to triggering a chain reaction during pullbacks. The market structure shows SK Hynix surged 7% in a single day, Micron followed with a 4%-5% rise, the Korean stock market rose 11% this week, and heavyweight stocks surged 15% over five days, confirming that funds are concentrating to drive a trend breakout in the memory sector. In terms of driving priority, the incremental demand of 1.2 zettabytes from data center expansion at the reasoning end ranks first, followed by the supply rigidity of advanced packaging capacity constrained until 2027, with short-term capital sentiment pressure ranking last. The upward scenario trigger condition is that $SNDK maintains a high-level narrow consolidation after a 20% surge in two days, and Micron can hold the lower support of the 4% breakout bullish candlestick. Variables to observe include declining turnover rate and narrowing range, with the invalidation signal being a break below the starting point of the rise. The downward scenario trigger condition is disorderly profit-taking at high levels, causing the price to break through the starting support of the two-day 20% increase. If the Korean stock market shows a large bearish candlestick after an 11% cumulative rise, it will trigger sector-wide synchronized decline. Overall, the failure point for judging trend continuation lies in whether it can stabilize above the support zone formed by this 7% to 20% increase. Breaking below this zone means the breakthrough structure based on tight supply and demand in 2027 is completely destroyed. The most important variables to observe in the next 7 days are the turnover volume of leading stocks in the memory sector and the effectiveness of support at the lower edge of high-level oscillations. #闪迪投资者日后股价大涨，长期目标待验证 #特朗普因TruthSocial付费数据流遭起诉 #标普收盘再创新高，8000点预期升温 More and more signs are starting to feel like the latter half of a bear market. The proportion of short-term BTC holders continues to decline. This phenomenon has appeared in the late stages of past bear markets. There are fewer short-term players and new funds are inactive. Market attention is decreasing. At the same time, chips are gradually settling into the hands of long-term holders. The hardest phase of a bear market is often not the daily big drops, but when the price falls so much that even the number of people discussing it decreases. The next step is when the proportion of short-term holders rises again from a low level. That would indicate that new participants and new demand are starting to enter the market again. 🌅 兄弟们，今早一睁眼，这个盘面属实有点意思。比特币大哥不慌不忙地跌了1.5%，ETH跟着滑了1.1%，连近期的当红炸子鸡SOL也只是勉强撑住，跌了0.2%。如果你只看这几个主流币，那感觉是风平浪静，甚至有点昏昏欲睡。但你眼光往旁边的犄角旮旯一瞟，嚯，好家伙，$EDEN一夜之间直接拉爆48%，$AEON也涨了20%。这感觉就像是什么？就像是你家小区里那家常年没人气的网红餐厅，突然排起了长队，你凑过去一看，发现屋里坐满的居然是另一波人，你压根不认识。我说句大实话，这种行情最容易让人心痒难耐。主流币歇菜，小币却像打了鸡血一样上蹿下跳，那种“牛市回来了”的味道实在太浓了，浓得像清晨巷口刚出锅的油条，闻着就让人想掏钱。但咱们老韭菜都清楚，这味道熟归熟，陷阱也往往就藏在里面。我根本不觉得这是什么真正的山寨季。你看，$EDEN涨了48%，跟第二名的涨幅差了28个点，这个差距说明啥？说明资金根本没打算撒芝麻盐，雨露均沾地照顾每个币，它就是集中火力，精准打击，有钱也只盯着几个小圈子玩，玩完一个换一个。这就好比一群大哥去夜场，不是给全场买酒，而是专挑几个穿着特别扎眼的姑娘送花篮，其他人只能干看着。 $SNDK From my understanding, as long as the AI development trend does not change, storage demand will not suddenly disappear. Hynix, as a key leader in the HBM field, still has mid-to-long-term performance logic, but how far the stock price can go ultimately depends on market sentiment, capital flow, and performance fulfillment. $SKHYNIX Additionally, I believe the logic behind the recent rise of SanDisk and Hynix is not entirely the same. SanDisk benefits more from the concentration of AI funds in the US stock market and improved earnings expectations, while the Korean stock has rebounded over 20% in the past ten trading days, partly due to repairing previous high-leverage sell-offs and overly pessimistic sentiment. In other words, one is more inclined towards industrial growth logic, while the other also includes valuation recovery. The most important question for the market to consider next is: Is this rise just a phase of repair, or the start of a new trend? Given the current situation, I personally will not chase the highs for now. Because short-term volatility is still significant, and market divergence is obvious. For companies with long-term optimism, it is indeed not necessary to pursue the lowest point, but during periods of intense emotional fluctuation, it is more important to wait for the trend and market environment to stabilize further. The hardest part of investing is not finding good companies, but finding your own balance between good companies and reasonable prices. #海力士扩产提速，资本开支能否兑现回报 #闪迪投资者日后股价大涨，长期目标待验证 #消费动能转弱，9月政策仍受通胀制约开篇：盘面现状现价约140美元，IPO解禁抛压、AI转型利好两股力量持续拉扯。最新SEC文件实锤：英伟达持有1.228亿股SpaceX股份，当前市值约170亿美元，成为公司第六大股东。资本绑定之外，双方敲定硬核业务合作，SpaceX的太空AI项目Starmind AI1，将全面采用英伟达Vera Rubin架构搭建轨道算力网络。核心矛盾：英伟达入股+太空AI故事打开长期天花板；短期股价承压、解禁抛压沉重，利好能否消化依旧存疑。一、支撑盘面的多头逻辑 1、重磅战略股东正式落地，信心背书极强英伟达直接跻身前六大股东，这笔股权源自年初对xAI的百亿投资，xAI被SpaceX收购之后完成股权转换，不是短期财务投机，属于长期产业层面深度押注。全球算力龙头真金白银重仓，向市场释放强烈的看多信号，能够一定程度对冲解禁带来的恐慌情绪。 2、业务深度绑定，拿到下一代AI硬件独家合作马斯克公开表态，SpaceX地面、太空AI算力优先选用英伟达方案。双方联合研发Starmind‑AI1，把Vera Rubin整套机架级算力系统部署到近地轨道，打造太空分布式AI数据中心，开辟地面之外全新的算力赛道 #Consumption momentum weakens, September policies still constrained by inflation 🚨 Consumers are starting to hit the brakes, but the Federal Reserve is still hesitant to ease off the gas, and the market in September may enter a phase of divergent trading! Weakening consumption momentum should originally be a dovish signal. Coupled with recent cooling in CPI and PPI, the logic for a rate hike in September is being gradually undermined. But the current U.S. economy presents an interesting combination: consumption is cooling down, yet inflation has not fully extinguished. So we cannot simply assume: weak consumption → rate cuts → BTC rises. What the Fed fears is not the slowdown in consumption, but the resurgence of inflation expectations. If this continues, the Fed will be in an awkward position: hesitant to raise rates, hesitant to cut rates, and ultimately forced to maintain high interest rates for longer. In this environment, the market will enter a typical "divergent trading" phase: For U.S. stocks, especially growth/tech stocks, there may be volatility plus structural divergence (the strong get stronger, the weak catch up with declines); Capital will favor: cash-flow-positive leaders, defensive sectors, and certain companies within the AI chain, rather than broad risk appetite expansion. Meanwhile, the crypto market ($BTC / $ETH) is more likely to experience volatility plus event-driven moves in the short term, rather than a clear trending market. The core of crypto remains liquidity expectations, with focus on these three lines: ① Whether consumption continues to weaken (determines if recession trades hold) ② Whether inflation expectations fall (determines if the Fed dares to pivot) ③ Whether U.S. Treasury yields truly decline (determines if liquidity is genuinely released) Only if all three weaken simultaneously: will the "easing trade" truly start, and both U.S. stocks and BTC will enter a risk appetite expansion phase. But if the current combination persists: consumption weakens, inflation expectations do not fall, and rates remain high, then the market’s essence is simple — the economy is deteriorating, but liquidity has not arrived yet. Do you think this market cycle is trading a "recession" or "stagflation"?Bitdeer Continues "Mine and Run" for 14 Consecutive Weeks — All 263.4 BTC Sold, Zero Holding Strategy Remains Unchanged --- 📊 1. This Week's Data: 263.4 BTC, Worth Approximately $17 Million On August 15, Nasdaq-listed Bitcoin mining company Bitdeer released its latest data: as of the week ending August 14, the company mined 263.4 BTC and sold all of it during the same period, resulting in a net increase of 0 BTC, continuing its zero Bitcoin holding strategy. Based on the current price range of about $64,000-$64,500, the sales this week are valued at approximately $16.9-$17 million. The previous week, Bitdeer sold 270.5 BTC, with recent production remaining stable. 🔍 2. From "Hodlers" to "Mine and Sell": At Least 14 Weeks of Zero Holding Strategy Bitdeer has implemented a zero net Bitcoin holding strategy since February 2026, lasting at least 14 weeks. Weekly production data shows steady output between 260-280 BTC since June. Previously, Bitdeer disclosed holding an inventory of 943.03 BTC and liquidated it all at once. In the first half of 2026, like peers Riot Platforms and CleanSpark, Bitdeer shifted to a mine-to-cash strategy, locking in cash flow amid the current high coin prices. 📉 3. Why This Approach? 1. Lock in cash flow and reduce price volatility risk Bitdeer's logic is simple: rather than betting on Bitcoin's future price movements, it immediately converts mined coins to cash, securing current revenue. Given BTC's ongoing fluctuation between $63,000 and $65,000, this is a conservative yet pragmatic financial strategy. 2. AI cloud business is the main capital focus Bitdeer is heavily investing in AI/HPC data center transformation — having secured a $4.7 billion, 16-year AI data center lease agreement in Tydal, Norway, with a global power capacity of 2,980 MW. Cash from coin sales supports this AI transition. 3. The "loss" backdrop in Q2 financials Bitdeer's Q2 report showed a net loss of $92.3 million, with adjusted EBITDA of $31.1 million. The accounting loss mainly stems from Bitcoin holding valuation changes — but with zero holdings, such "paper losses" would not occur. 📉 4. Market Reaction: Reflected in Financials, Pricing Logic Unchanged After Bitdeer's Q2 report, the stock price did not experience sharp fluctuations, indicating the market has fully priced in the zero holding strategy. The $4.7 billion AI data center deal in Norway and AI cloud business are the variables attracting market attention — Bitdeer is evolving from a "mining company" to a "mining plus AI infrastructure company." 💎 5. Summary Bitdeer's "mine and sell" strategy has lasted at least 14 weeks, with a stable weekly sell pressure of about 260-280 BTC worth roughly $17 million, becoming a structural supply the market continuously absorbs. This does not change Bitcoin's long-term trend but explains why BTC struggles to rise around $64,000 despite ongoing ETF inflows — besides Strategy's selling and Jump Crypto's transfers, miners are steadily offloading. For Bitdeer, Bitcoin is merely a "product," not a "reserve asset." Whether the zero holding strategy continues until Bitcoin breaks $70,000 will be a key window to observe its strategic resolve. $BTC The storage sector is now playing "hunger marketing"—except this time it's real hunger. Micron executives have said outright: data center storage shortages will last at least until 2027. Omdia is even more severe: AI demand has already exceeded chip manufacturing and packaging capacity; bottlenecks like HBM and advanced packaging won't be resolved before 2027. To translate: AI, this "big eater," is consuming storage to the point of stockouts, and the shortage notice has already been extended to 2027. Just look at the market: SanDisk rose 20% in two days, Western Digital and SK Hynix surged 7% in a single day, Micron and Seagate also gained 4%-5%. Even the Korean stock market went crazy—KOSPI rose 11% this week, ending a seven-day losing streak, with Samsung and SK Hynix up more than 15% in five days. Korean investors: Seven years! Do you know how I've lived through these seven years?! The longer story is: after AI moves from training to inference, massive daily conversations and intelligent agents generate tokens nonstop, and every token needs storage space. SanDisk says that by 2030, the enterprise data center flash market will reach 1.2 zettabytes—what does 1.2 zettabytes mean? It means copying all the world's hard drive data several times over. The biggest difference from 2021 in this cycle is that demand is real and sustained. Training requires GPUs, inference requires storage; the hardware chain's resources are consumed piece by piece. But with such rapid short-term gains, the old saying applies: chasing highs feels good for a moment, but corrections are a crematorium. $SNDK Currently, global memory chip leader SK Hynix is accelerating its capacity expansion at an unprecedented pace. Financial reports and publicly available market data show that SK Hynix's capital expenditure (Capex) for 2026 has been significantly raised to a high level of 40 trillion KRW (expected between 40 trillion ~ 50 trillion KRW), and its actual cash expenditure on tangible assets in the first half of 2026 has surged 72.7% year-on-year to 18.33 trillion KRW. In addition, its board recently approved a medium- to long-term special investment plan for new plants totaling 54.3 trillion KRW (such as Yongin Y2, Cheongju M17, etc.). Although some quarters (such as Q2 2026) were affected by business structure and profits were short-term disrupted by friction costs from high-end capacity transitions, the high premium of HBM (HBM3E and its evolution to HBM4) has kept its overall profitability at historic highs. R&D expenses nearly doubled year-on-year in the first half (reaching 6.04 trillion KRW, a 98.4% year-on-year increase), showing that it is not only spending heavily on equipment but also fiercely breaching technical barriers to ensure technological gaps. Capacity built through large-scale capital expenditure is mostly directly connected to long-term orders from North American hyperscalers and AI chip giants. This deeply intertwined business model allows massive capital expenditures to be gradually stabilized and recovered over the next few years through high cash flows. South Korea's semiconductor industry is highly dependent on global supply chains (such as semiconductor equipment from the US, Japan, and Europe).Revenue Halved, Net Loss of $70.1 Million—Bithumb's Darkest Hour Also Reflects the Korean Crypto Market --- 📊 1. Key Data: From 55 Billion Won Profit to 108.7 Billion Won Loss On August 15, Bithumb released its Q2 2026 financial report: First half (YoY): · Revenue: 168.8 billion KRW (approx. $109 million), down 48.7% · Operating profit: 14.9 billion KRW (approx. $9.6 million), down 83.4% · Net profit: loss of 108.7 billion KRW (approx. $70.1 million), compared to 55 billion KRW profit in the same period last year Second quarter (YoY): · Revenue: 86.3 billion KRW (approx. $55.7 million), down 35.8% · Operating profit: 12.1 billion KRW (approx. $7.8 million), down 44.0% · Net profit: loss of 21.8 billion KRW, compared to 22 billion KRW profit in the same period last year 🔍 2. Reasons for Loss: Three Blows Striking Simultaneously Blow One: Plummeting Trading Volume—In the first half of this year, the combined trading volume of Korea's five major KRW exchanges (Upbit, Bithumb, Coinone, Korbit, Gopax) was about $366.58 billion, down 54.6% YoY. In May, crypto market trading volume as a percentage of KOSPI dropped from 11.29% in January to about 2%. Blow Two: Mass Exodus of Korean Retail Investors—Under high interest rates, Korean retail investors are massively withdrawing from the crypto market, shifting to AI and semiconductor stocks. The five exchanges' average daily trading volume fell from about $2.82 billion last year to about $305 million, a decline of approximately 89%. Blow Three: Digital Asset Impairment Losses—Of the 108.7 billion KRW net loss in the first half, digital asset valuation losses were a significant component. During the market downturn, the book value of digital assets held by Bithumb sharply decreased, further widening the losses. 📉 3. Market Landscape: Upbit "Winner Takes All," Bithumb Falling Further Behind From July 1 to 27, Bithumb's trading volume dropped to 4.71 trillion KRW, with market share falling from 30.7% to 27.1%. Meanwhile, Upbit's market share rose from 62.3% to 67.4%, widening the gap to 40.3 percentage points. In a bear market, money doesn't disappear; it just flows to platforms that can best retain liquidity. 📋 4. Bithumb's "Three Self-Rescue Strategies" 1. IPO Sprint: Officially announced IPO plans in May this year, targeting listing by 2028. Preparing for K-IFRS conversion in 2026 and submitting the listing pre-review application in 2027. However, with current performance, the IPO valuation may not look attractive. 2. Betting on Compliance: Improving compliance systems around the expected "Digital Asset Basic Act" to be introduced this year, and proactively preparing for opening the digital asset market to corporations. 3. Product Differentiation: Using AI-driven trading convenience and investment information services to cope with the sluggish trading market. 💎 5. Summary Bithumb's halved revenue and $70.1 million net loss expose not only its own operational difficulties but also reflect the overall contraction of the Korean crypto market. As retail funds massively flow from crypto to AI and semiconductor stocks, exchanges are the first to be hit. Upbit has absorbed most existing users with stronger liquidity, while Bithumb is falling further behind. The most ironic part: Bithumb is losing money and market share yet plans to IPO in 2028. Whether this "bear market survival story" succeeds depends on whether it can endure the coldest winter ahead before the "Digital Asset Basic Act" is implemented in 2027. $BTC $SNDK has already stabilized at 1600. On the day of the investor event, it rose 13.7%, a gain stronger than many altcoins. Currently, liquidity is basically flowing into US stocks, with global capital continuously investing. AI hardware is supporting valuations, and AI storage demand remains. $SKHY has also risen 15 points since August, with a massive 540 trillion KRW expansion order, and it also collaborates with $NVDA Nvidia. There are indeed concerns in the market about overcapacity, but I believe the AI trend continues and storage demand remains. As the absolute leader in HBM, performance is not a concern; the key is to watch market reaction. #HynixExpansionAccelerates, can capital expenditure deliver returns The recent rise in SanDisk and SK Hynix is mainly driven by AI demand, performance, and institutional funds. The Korean stock market rebounded 22% in 10 days, which I think is more about repairing the golden pit caused by previous high-leverage liquidations, a different nature. #KoreanStocksReboundOver22InTenDays, chip stocks lead the rise Will it continue to rise, or has it peaked? This is the question we need to consider. At present, I dare not intervene. Good companies can make money whenever you buy, I understand this principle. But the current volatility is too high, so I will wait for the market to stabilize a bit before acting. Stocks with long-term potential shouldn’t fluctuate so wildly every day. High volatility now itself indicates significant market divergence. When the divergence converges, the direction will naturally emerge. #闪迪投资者日后股价大涨，长期目标待验证 American consumers are starting to hit the brakes, and the Federal Reserve's toughest moment may be coming Inflation has finally dropped, but another bigger problem is emerging. The biggest support for the U.S. economy in the past — consumption — is showing signs of cooling. The latest data shows that U.S. retail sales in July fell by 0.6% month-over-month, not only below the market expectation of a 0.1% increase but also the most significant decline in over a year. Meanwhile, the consumer confidence index also weakened, with the University of Michigan Consumer Sentiment Index falling to 51.0, below market expectations. This indicates an important change: The U.S. economy is no longer simply facing "high inflation," but rather "inflation has not completely disappeared, yet consumption momentum is starting to weaken." This is precisely the most difficult situation for the Federal Reserve to handle. Over the past two years, the Fed has maintained high interest rates to suppress demand, cool the economy, and thereby control inflation. Now, it appears this strategy is taking effect. The previously released July CPI and PPI data show that price pressures continue to ease, with inflation cooling simultaneously on both the production and consumption sides. At the same time, retail data is slowing down, indicating that the high interest rate environment is gradually transmitting to households. But problems are also emerging. If consumption continues to weaken and the Fed keeps interest rates high, it may further suppress economic growth; if policy is relaxed too early, inflation could rebound. Therefore, the core of the September meeting is no longer just about inflation numbers but about finding a balance between "controlling prices" and "protecting the economy." Many investors are now focused on rate cut expectations, but I believe the market is actually trading a shift: The U.S. economy is moving from "demand overheating" to "demand cooling." Previously, the market worried whether the Fed would continue raising rates due to persistent inflation. Now, the new question is whether the Fed will be forced to pivot early because of increased pressure from consumption and employment. However, it is important to note that a decline in consumption does not mean the U.S. economy is immediately entering a recession. American consumers still show resilience. Currently, the consumption slowdown is more concentrated in some discretionary sectors, such as automobiles and online retail, while services like dining still maintain some growth. This indicates the U.S. economy is not suddenly stalling but gradually slowing down under a high interest rate environment. From an asset perspective, this change will bring new impacts. If inflation continues to decline in the future while consumption and employment weaken further, the Fed's policy space may open up. In this case: The dollar may come under pressure; U.S. Treasury yields may fall; Gold may continue to benefit; Risk assets will reprice expectations for improved liquidity. For BTC, the logic is similar. One of the key drivers of Bitcoin's rise in recent years has been changes in global liquidity expectations. If the market starts to believe the Fed is entering a policy pivot phase, risk appetite may rise again. Conversely, if inflation rebounds and the Fed maintains high rates longer, all high-valuation assets will face renewed pressure. My view is that the market is now entering a very critical observation phase. Previously, everyone focused on CPI to see when inflation would come down. Going forward, it is more important to see whether American consumers can continue to hold up. Because consumption accounts for a large portion of the U.S. economy, changes in consumer confidence and actual spending will directly affect corporate earnings expectations. This is why in the coming months, the direction of U.S. stocks, the dollar, gold, and BTC will not be determined by a single data point. What truly determines the trend are three signals: Whether inflation continues to decline; Whether employment continues to weaken; Whether consumers can maintain purchasing power. If these three directions change simultaneously, the Fed's policy cycle may see a real turning point. What deserves the most attention now is not whether there will be a rate cut in September, but that the U.S. economy is moving from a high-speed running phase into a new balanced phase. And the capital markets are pricing in this change in advance. $OKB $DOS $ETH #消费动能转弱，9月政策仍受通胀制约 Currently, there are three companies capable of producing HBM. In terms of technology, SK Hynix is the strongest; they were the first to develop it, followed by Samsung, and the weakest is MU, because MU mainly focuses on edge computing devices, like clients such as Apple. But why do I have more confidence in the technically weakest $MU rather than $XSKHY? Because of the Korean chaebols. Actually, I think East Asians have similar personalities. If you buy SK Hynix stock, they won't really respect you as a shareholder. Moreover, SK Hynix has a parent company above it, which can lead to conflicts of interest. Micron is an American company, and I trust American rule of law and American stocks more. The management also respects shareholders. Therefore, I am long on the technically weakest MU rather than SK Hynix. The market thinks the same way, giving MU the highest valuation, not SK Hynix. #海力士扩产提速，资本开支能否兑现回报 The underwater limit was already 13,000, and I realized the biggest pain was never losing money, but the "if" during the review. Have you ever had that moment when you woke up in the middle of the night, replaying the same deal in your mind, asking yourself what you did wrong? That was how I was last night. Looking at SanDisk's chart, Hynix isn't that strong, so why is SNDK surging so fast? Later, I realized that market trading has never been about the fundamentals of the present, but about the "story of the future." SanDisk Investor Day sets long-term targets, so funds are already injecting expectations in advance. This pricing logic is exactly the same in the crypto market. I remember every time BTC surged, on-chain data showed retail investors chasing and whales pulling out. But prices still go up—why? Because what everyone buys is the "next narrative," not the current on-chain activity. When the market starts valuing BTC with "future revenue expectations" instead of looking at actual on-chain settlement volume, this itself signals a shift from rationality to imagination. Where is the most vulnerable link now? On leverage. Once the funding rate for perpetual contracts remains positive, long crowding will be maxed out. SanDisk's "meeting-driven" rally corresponds to ETF inflow data, Fed speeches, and some influencer making orders in crypto—the shock of events comes quickly and fades just as fast. If I had to do it again, what would I do? They first ask themselves: Is this rally being bought by spot stocks or by contracts? Spot trading is about consensus, while contracts are driven by sentiment. Emotions come quickly and go even faster. The bullish path is: if BTC can hold steady,Liquidation data reveals the key bullish and bearish watershed for BTC, while ETH lacks options capital of the same scale for competition On-chain liquidation monitoring data shows a clear price threshold for Bitcoin: if the price dips near 62000, a large number of long positions will be liquidated; conversely, breaking above 64000 USD will result in a concentrated squeeze of accumulated short positions. In comparison, it is obvious that $ETH's options and contract liquidation concentration is far lower than BTC's, which is a major feature of the recent market: BTC's price swings often trigger concentrated pulse movements, while Ethereum tends to follow passively and struggles to independently establish a trend. The underlying logic is that $BTC has already been incorporated into asset allocations by a large amount of traditional capital, with deep participation from derivatives institutions; $ETH still relies more on native crypto community funds for competition. In the short term, do not simply assume that when BTC starts to rise, Ethereum will necessarily follow suit. In the current market's existing competitive environment, the strength differentiation among mainstream assets will continue to play out ETH discussion has slowed down, let's first look at the denominator of this tone This round of ETH numbers has a sense of direction, but I am more concerned about the sample size. OKX Onchain OS recorded 17 mentions in one hour at 11:00 on August 15, with 47% bullish and 6% bearish, and the discussion speed is about 0.88 times the 24-hour hourly average. A few concentrated reposts can significantly rewrite the ratio, so "bullish clearly dominant" can only describe this batch of texts and cannot be equated with how much capital is betting in the same direction. Regarding sources, X had 17 mentions, news had 0 mentions, so we also need to watch if the same news is being repeatedly spread. Next, we will see if the tone can be maintained after expanding the sample, then cross-verify with trading volume, funding rates, and on-chain activity, which is more reliable than drawing conclusions based on a single percentage.#OpenAI与Anthropic估值竞赛升温 Damn! OpenAI and Anthropic, two cash-burning giants, are using trillions in paper wealth to swallow up the entire liquidity of risk assets. Anthropic has completely ridden over OpenAI. The $65 billion financing round in May pushed its valuation to $965 billion, surpassing OpenAI’s $852 billion for the first time. Now with an annualized revenue of $47 billion, the proportion of enterprise clients has surged from single digits to over 34%, and 70% of Fortune 100 companies are using Claude. Investors are already privately shouting that the October IPO will open at $2 trillion, with some even calling for $3 trillion. Recently, Reuters revealed their internal forecast that revenue will hit $190-200 billion by 2028, and bankers have started applying multiples based on that. This isn’t just going public; it’s treating Wall Street like an ATM. OpenAI is still clinging to the “900 million weekly active users” face value, but the number of actual paying customers is pitifully low, and the old trick of losing two dollars to make one dollar has been played all along. In Q1 alone, they lost $2-3 billion. The CFO and Altman nearly fought in the boardroom over when to go public. Some have already called it a “charity AI company.” Now annualized revenue has just reached over $40 billion, valuation is stuck at $852 billion, and the IPO target is firmly fixed at $1 trillion. After SpaceX went public and the market shook a bit, they chickened out and pushed the IPO timeline to 2027. Together, these two have already sucked more than $200 billion in real money out of the market. This money could have flowed into crypto and risk assets, but now it’s all siphoned off by AI unicorns. SpaceX, OpenAI, and Anthropic combined have valuations exceeding $3.6 trillion, all rushing to the public market. Institutional funds have no appetite left to share a piece of Bitcoin. In this era of zero-sum competition, the bigger the whales’ appetite, the fewer scraps are left for BTC. Professional analysts on X also believe: “It doesn’t matter which AI giant rings the bell first; what matters is whether the valuation is stable. If stable, the tech sector will thrive and Bitcoin will benefit; if it crashes, the entire sector will be repriced, and Bitcoin will shake too.” Some have expressed concerns: when a round of AI financing exceeds the weekly ETF inflows into the crypto market, it’s a naked short-term bloodletting for BTC. The cycle of financing is so intense that cloud providers are both investors and clients. Once sentiment reverses and the AI bubble bursts, risk appetite will collapse, and the crypto market will be the first casualty. But some see it more simply and far-sightedly: Wall Street is pricing computing power at the trillion-dollar level with real money, essentially confirming that “computing power is the new oil.” Bitcoin, as the most original and hardcore expression of computing power, will only get stronger in the long-term narrative, not weaker. Short-term bloodletting is real, but long-term elevation is also real. Actually, these two brothers aren’t fighting; they’re jointly locking all capital market attention and liquidity onto the AI track. In the short term, the crypto market is the scapegoat, with funds drained and narratives diverted. In the long term, if their valuations stabilize and the financial attributes of computing power are confirmed, Bitcoin could actually take off. When October comes, if Anthropic really pulls off a $2 trillion IPO, the tech world will continue to thrive, and Bitcoin will benefit a bit; if it crashes, the whole market will bleed, and no one will escape. Every time Bitcoin hits a bear market bottom, when you open the global candlestick chart, the pattern always makes you feel it will drop further. In 2023, when it was at 15,000, many people said it would go down to 8,000. In 2018, at 3,000, many said it would drop to 1,000. I didn't experience 2015, but if you look at the chart, wasn't the 2015 situation hanging "in the sky"? Now many people look at the pattern and say it will drop further. Actually, the 57,000 in June this year was the lowest point of this cycle; you just didn't buy then. #消费动能转弱，9月政策仍受通胀制约 스타십 이후 SPCX, 해제 물량과 가격 방어 사이에서 갈림길 월간 단위 토큰 해제가 겹친 가운데, 왜 시장은 100달러 아래를 확신하면서도 반등을 허용했을까? 원문에서 확인된 핵심 사실은 세 가지다. 첫째, SPCX 가격이 한 달 기준 하락 50%, 상승 40%라는 극단적 변동성을 기록했다. 둘째, 월간 단위 토큰 해제가 지속 중이며, 셋째, 엘론 머스크의 공개 발언이 가격 급등을 촉발한 정황이 포착됐다. 이는 단순 변동성 확대가 아니라, 해제 물량이 시장에 풀리는 구조적 공급 압력과 특정 발언이 맞물린 이벤트성 랠리다. 여기서 시장 구조를 보면, SPCX는 전통 주식 시장에서 거래되는 자산이지만 최근 크립토식 변동성을 보여준다. 이는 해제 물량을 소화해야 하는 매도 압력과, 머스크 발언에 베팅하는 매수 세력 간의 힘겨루기로 해석할 수 있다. 100달러는 단순 심리적 지지선이 아니라, 해제 물량의 손익 분기점이자 숏 포지션의 청산 기준선일 가능성이 크다. 즉, 가격이 100 아래로

Orbit Topic Sitemap