钞能力玩家

"From AI to the endpoint, can NVDA still be chased?" Micron's revenue surged 196%, with a gross margin of 75%. Dell's AI servers increased by 757%, with full-year guidance raised to 60 billion. NVIDIA launched RTX Spark, packing 1 PFLOPS of computing power into a laptop. Three signals combined: AI has spread from data centers to our desktops. The chain logic remains unbroken; endpoint demand is just beginning. Last night NVDA rose 6.26%, with a market cap surpassing 5.43 trillion. OKX perpetual contracts are tradable 24/7, allowing crypto funds to follow seamlessly. But at this level, short-term sentiment is already very stretched. The bullish trend remains unchanged, but don’t go all-in at the peak of sentiment. A pullback is a buying opportunity; better to wait than to chase the tail. #美股洞察：英伟达发布新款AI PC芯片

"Rocket Body vs AI Soul: What Makes SpaceX Expensive?" $75 billion in financing, a $1.75 trillion valuation. But looking at the financial report, the truth isn’t that romantic. Starlink Q1 revenue is $3.257 billion, operating profit $1.188 billion, EBITDA margin 63%—this is a real cash cow. In contrast, the AI segment’s Q1 revenue is only $818 million, operating loss $2.469 billion, with $7.7 billion capital expenditure all spent here. In other words, the real breadwinner is satellite internet, but the market’s premium is entirely bet on the AI narrative. Looking at the Cursor acquisition: pushed 30 days after IPO, $60 billion to buy an AI programming tool company, or else pay a $10 billion breakup fee. What does $60 billion mean? SpaceX’s entire AI segment revenue for 2025 is only $3.2 billion. This money isn’t buying Cursor’s current revenue but a strategic entry into the AI programming track. Which has more weight, AI acquisition premium or the rocket main business? The answer is already written in the pricing. A pure rocket company isn’t worth $1.75 trillion, but the combination of “Starlink cash flow + AI burning money for the future + space monopoly” is. The pre-IPO perpetual contract is already priced 34% above the target valuation, implying a market value of about $2.3 trillion. Buying SpaceX means buying Starlink’s cash flow, but paying the ticket price for the AI blueprint to interstellar space. Is it worth it? It depends on how much you’re willing to pay for the "story". #SpaceX IPO新增增发预警

"Trump says within a week, oil prices will listen to Iran" Trump said "an agreement will be reached within a week," and the Strait of Hormuz will be navigable. But the Revolutionary Guard responded directly with missiles: the strait "remains closed." The market didn't hesitate—Brent crude broke through $97 intraday and closed up 4.2%. Talks on the surface, blockade in action. Capital chose the latter. Within the $95 oil price, there is already a $10-15 geopolitical premium embedded. My approach: go long with a small position, set stop-loss. If the agreement really materializes, the premium will disappear; if the situation escalates, oil prices could hit $110. Trump's optimistic statements seem more like expectation management to lower oil prices, not intelligence assessment. In the coming week, either the premium evaporates or continues to expand. Don't bet on Trump's words, bet on Iran's missiles and the course of their cargo ships. #美伊交火：特朗普称“小插曲”

"Two rules are in conflict, how will UMA rule this time?" Strategy sold 32 $BTC from May 26-31, marking the first sale in four years—but the announcement was only posted on June 1. On Polymarket, a market with $50 million in trading volume, the core bet was whether the sale happened before May 31. Two interpretations clashed: counting by sale time, the vote should be "Yes"; counting by disclosure time, the vote should be "No." Polymarket itself couldn’t make the call and passed the decision to UMA token holders to vote. Where’s the loophole? The rules don’t specify whether information disclosure or on-chain transactions take precedence. This is what UMA needs to decide. So how will the vote go? Ultimately, it depends on UMA’s voting structure. Data from the past three years shows that 9 major holders control nearly half the voting power—these 9 addresses almost always vote in sync and always win. What’s more noteworthy is a survey: in 72% of controversial votes, token holders themselves bet on the outcome. At that point, who even cares about the facts? So rather than UMA arbitrating the facts, the ruling power has always been concentrated in the hands of a few. The final result might still be "No," because that aligns better with the established interests of the major holders. #Strategy披露上周出售32枚比特币

"$100,000 Bet: How Long Can HYPE Stay Hot?" $HYPE just broke a new high at $74, and Arthur Hayes made a $100,000 bet with Multicoin's Samani—betting that HYPE will outperform all tokens in the top ten by the end of the year. These two aren't ordinary KOLs; one is the founder of BitMEX, the other a top fund partner. Their reputations are directly tied to the price. This bet essentially bought HYPE six months of attention. Short-term sentiment has been ignited, and institutions are still scrambling to accumulate and lock in. But can the hype last until the end of the year? Unlikely. Unless there's a dramatic turnaround—like HYPE briefly overtaking $SOL, or Samani posting another provocative tweet. Otherwise, the market will gradually become numb and eventually return to fundamentals: protocol revenue, buyback flywheel, liquidity. The $100,000 bought exposure, but whether HYPE can capitalize on this traffic depends on itself. #HYPE连创新高：ETF持续净流入

"Anthropic files first, who's really panicking?" Anthropic has secretly filed, valued at $965 billion, surpassing OpenAI's $852 billion for the first time. Annualized revenue is $47 billion, expected to turn profitable in Q2 for the first time. In contrast, OpenAI loses $1.22 for every $1 earned in Q1. One is approaching profitability, the other is still burning cash to fuel growth. The two are taking very different business paths. Almost all of Anthropic's core revenue comes from enterprise clients, with B2B gross margins rising from 38% a year ago to over 70%, fully reflecting the profit margins of a mature high-end SaaS company. OpenAI's Q1 revenue was $5.7 billion, but weekly active user growth has basically stalled, and paid conversion rates are very low. Revenue growth is clearly slowing, and the venture capital community sees the ceiling. But filing first also carries risks. OpenAI can now watch Anthropic reveal its cards to institutions first, stumble through pitfalls, and then adjust its own pricing strategy. So on IPO pricing day, who has the advantage? Market investors have a simple logic: they are willing to pay a premium for "profitable AI companies" but won't wait indefinitely for "companies that keep burning cash while telling stories." At least for now, the balance of victory has already tilted. #Anthropic递交招股书：正式启动IPO $OPENAI $ANTHROPIC

$ALLO is approaching its value defense line, with 0.1650 as the touchstone. Chips are continuously changing hands; after washing out floating chips, the foundation becomes more solid. RSI has quietly slipped into the oversold zone, indicating short-term selling pressure is nearly exhausted. Volume is narrowing, which is not a bad sign— this shows panic is ebbing, and the market is waiting for direction. MACD is still below zero, but bearish momentum is weakening. At this time, no need to be aggressive; keep a close eye on 0.1650. If it can hold steady on low volume, that’s the most concrete signal the market can give. #波动雷达：币种异动观察

"NVIDIA up 2%, Intel down 6%: Same question, two different answers" NVIDIA launched the RTX Spark, a fully self-developed Arm architecture PC chip integrating the Blackwell GPU, delivering 180~200 TOPS edge AI computing power, paired with 128GB unified memory. This marks NVIDIA's first entry into the PC processor market as a "major player," directly taking market share from Intel and AMD. Intel's stock dropped over 6% that day, falling to $108. AMD and Qualcomm also fell between 3% and 6%, while NVIDIA itself rose about 4%. This is the market voting with its feet—the pricing logic of PC chips is being rewritten by AI. Digging deeper, Intel's forward P/E ratio is as high as 156, barely profitable. NVIDIA's P/E ratio is only 27, with a PEG ratio of just 0.68, surprisingly cheap for an AI stock. When the market revalues Intel using NVIDIA's valuation logic, if Intel doesn't fall, who will? One is advancing AI-native architecture, the other clings to the old x86 legacy. This divergence will at least continue until Intel's "Crescent Island" inference chip ships and delivers results. Until then, the PC chip story is no longer "whose CPU is faster," but "whose chip can run AI." #美股洞察：英伟达发布新款AI PC芯片

"SpaceX's Bitcoin: Reserve or Ammunition?" 18,712 BTC, unrealized gains doubled, market value over $1.4 billion. The S-1 filing just disclosed this figure, twice what on-chain tracking showed. But the prospectus added a new risk warning: a potential large-scale equity issuance. The option to acquire Cursor will be clear 30 days after the IPO. A $60 billion deal, paid either in equity or cold hard cash. Once cash flow pressure mounts, the $1.4 billion in Bitcoin becomes the most convenient ammunition. Unrealized gains at 119%, the finance department can’t ignore this accounting. The problem is the market hasn’t priced in this risk at all—everyone’s discussing "SpaceX holding coins to pump," no one’s thinking "Will SpaceX dump?" Strategy sold 32 BTC, MSTR dropped 9% intraday. If SpaceX really considers this move, the scale is on a different level. Against a $2 trillion valuation IPO, $1.4 billion in BTC is just a chip on the table. When they move is the real question. #SpaceX IPO新增增发预警 $BTC $SPACE

"A 'minor incident'? The market just slapped Trump in the face" Trump called it a "minor incident" and said oil prices would "plummet like a rock." But as soon as he finished speaking, Brent crude surged to nearly $97, with US oil jumping 5.5% and Brent up 4.2% the previous trading day. Where does this gap come from? Since May 29, US missiles have struck 5 commercial ships and forcibly altered the course of 116 vessels. A US military base was attacked, and an MQ-9 "Reaper" drone was destroyed, costing $30 million each. The Iranian Revolutionary Guard directly retaliated against US and Israeli ships with cruise missiles. Is this a minor incident? There is no ceasefire between the parties, and the Strait of Hormuz is effectively "blocked." Analysts warn that if the strait is completely closed, Brent crude could break through $110 directly. So the market isn’t doubting Trump—it’s telling him through real actions: before a ceasefire takes effect, every bombed cargo ship adds real money and premium to oil prices. And Trump’s casual talk about oil prices being like the weather is because the US doesn’t lack oil. For European and Asian countries heavily dependent on oil imports through the Strait of Hormuz, this is neither a "minor incident" nor will oil prices drop like a rock in the short term. #美伊交火：特朗普称“小插曲”