Orbit Post Sitemap
This is a major development for the AI infrastructure narrative.
The key takeaway is that NVIDIA is effectively helping turn AI compute into a financeable infrastructure asset. If the announced platform can mobilize $500B+ of third-party capital, the implications could extend far beyond NVIDIA itself:
🏗️ AI infrastructure: More capital for data centers, power, networking, and compute capacity.
🏦 Institutional participation: Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR bring enormous financing capacity.
⚡ “AI factories”: Jensen Huang’s framing suggests compute is increasingly being treated like critical infrastructure rather than simply hardware.
💰 Capital cycle: Large-scale financing could accelerate AI infrastructure deployment and potentially create opportunities across the broader AI supply chain.
📈 Market implication: The AI boom may increasingly become a capital-intensive infrastructure investment cycle, rather than just a semiconductor story.
The most interesting part is the shift in mindset: Wall Street isn't just investing in AI companies—it is increasingly preparing to finance the physical infrastructure required to run AI at massive scale.BTC & ETH ETF Inflows Return: Institutions Are Buying — But Fed & Hormuz Hold the Key
The crypto market is entering a critical macro window. Institutional capital is returning, with U.S. spot Bitcoin and Ethereum ETFs attracting roughly $1.1 billion in combined net inflows over the past week. Yet $BTC and $ETH remain volatile as investors await the next catalyst.
The key question is whether ETF demand can overcome macro pressure.
All eyes are on U.S. CPI and the Federal Reserve. Softer inflation could strengthen expectations for Fed easing, lower yields and renewed risk appetite—conditions that would favor $BTC and $ETH.
But another major variable is the Strait of Hormuz.
Uncertainty over its reopening has pushed oil prices higher, reviving inflation concerns. Oil surged around 5% amid renewed uncertainty over U.S.-Iran negotiations.
This creates a critical macro battle:
ETF inflows = institutional demand.
Softer CPI = potential Fed easing.
Higher oil from Hormuz = renewed inflation risk.
If CPI comes in softer while oil pressure eases, global liquidity could improve. $BTC may benefit first, followed by $ETH as institutional adoption, staking and tokenization expand.
Beyond the majors, $SOL remains a key asset if risk appetite returns, while $OKB could benefit from stronger exchange activity and recovering liquidity.
The market is not simply waiting for a breakout. It is waiting for confirmation that macro conditions are turning supportive.
A dovish Fed outlook + sustained ETF inflows + easing Hormuz tensions could create a powerful setup for the next crypto expansion.
But hotter CPI + higher oil + geopolitical uncertainty could keep investors defensive.
For now, the most important signal may not be today's price.
It is where institutional capital is positioning before the next macro catalyst.
If you find these insights useful, follow me to keep tracking, analyzing and discussing the hottest developments across crypto and Wall Street.
#BTCETHETFFlowsDiverge
#HormuzDealUnresolved
#CPIToResetFedBets
$BTC
$ETH
$SOL 500 billion vs. 20 billion! Understanding the harsh truth 📉📈 of the second half of AI through financing paths
Both focus on AI infrastructure, so why are the treatment rates so vastly different between Nvidia and Intel? The amount of information in this chart is worth savoring for all tech investors.
Core Differences Breakdown:
✅ NVIDIA (NVDA): The game-maker
Action: Jointly established a financing platform with BlackRock and Goldman Sachs.
Essence: asset-light operations + financial leverage. It's no longer just a company selling GPUs—it's becoming the 'central bank' of the AI era. By empowering customers (lending money to customers), it locks in future orders. This is a long-term double boon for the stock price (performance + valuation boost).
⚠️ Intel (INTC): The Breaker
Action: Raised $20 billion through common stock issuance.
Essence: Heavy asset gambling. This is betting shareholders' money on future advanced processes. Although the subscription boom shows institutions are still willing to give it opportunities, it also means huge capital expenditure pressure. If process technology can't catch up in the next two or three years, that 20 billion yuan will be a heavy burden.
Conclusion:
The market is repricing "financing ability." In the AI money-eating game of the money-eating beast, NVIDIA is making money from others' money, while Intel is hollowing out its own resources to survive.
For those holding these two tickets, your strategies may diverge. #AI基建融资升温, Nvidia and Intel are taking divergent paths Tesla 2026 Q2 delivered a "highest revenue ever + nearly worst operating quality ever" earnings report. Revenue was $28.24 billion (+26%, record), but operating profit was $398 million (-57%), operating margin 1.4% (4.1% same period last year), free cash flow -$1.09 billion (first quarterly cash burn since early 2024). More notably, the H1 panorama: H1 2026 revenue $50.6 billion (+21%), operating profit $1.34 billion (vs $1.32 billion last year), net profit $1.59 billion (vs $1.58 billion last year) — 21% revenue growth resulted in 0% profit growth. $TSLA Anthropic poured $9 billion into Riot to buy hash power, and BTC miners are switching careers to become AI contractors—this is even more important to watch than market fluctuations.
BTC showed no signs of anger this hour: 64,071, 24h -1.86%, volume still down 87.8%, Fear stuck at 29, standard zombie market.
True Dark Thunder dropped -16.6% in GRVT: 24h, OKX's biggest dropper. A perp DEX token crashed so quickly, unrelated to stagnation—it's its own risk.
Miners have two arguments about AI orders: the good side is more income and less forced to sell BTC to cover electricity bills; the negative side is that AI giants are directly involved, competing for power and graphics cards in the future, making miners' costs even more competitive.
One thing that can be taken away: Don't just focus on BTC prices, but watch the progress of mining companies' "computing power transformation"—whichever company really secures an AI order will see less selling pressure. This is a new variable on the supply side.
AI giants are personally buying mining machine hashrate—does BTC have an extra backer, or a rival to power grabbing power? Comment says your judgment (shallow comment hidden, only pre-posted with reasons).
—— Veteran TCM Doctor on the Chain · Pulse Diagnosis Every Hour · 2026081116 · Computing power variable
Crypto assets carry high risk. This article does not constitute investment advice and reflects purely personal opinions.
#OKX星球 $BTC #AI算力 #矿工叙事8.9 billion in revenue, profits multiplied 135 times, and the stock price fell 47%—SanDisk's investor day, was it a lifeline or just another blow?
Quarterly revenue was $8.965 billion, up 372% year-over-year.
Adjusted earnings per share were $39.25, 135 times the $0.29 a year earlier.
Gross margin was 84.6%.
Eight long-term agreements lock in $93.9 billion in guaranteed revenue for the future.
A $14 billion share repurchase program.
And then?
The day after the earnings report was released, the stock once fell more than 13% intraday.
As of the close on August 10, SanDisk was at $1,238—a 47% drawdown from its June all-time high of $2,354.
The better the performance, the harder the drop.
Does this script sound familiar?
"Didn't you say there was an explosion in demand for AI storage?"
"Didn't you say NAND supply is short-lived?"
"Didn't you say the data center business grew 1298% year-on-year?"
The market said: I know. But I don't care anymore.
Breaking down the financial report, where did the problem lie?
First, the value of the performance is questionable.
SanDisk's management admits that of the 51% quarter-on-quarter revenue increase this quarter, only one-third came from increased shipments, while the remaining two-thirds came from NAND price hikes.
In other words, this growth is not driven by demand, but by price increases.
How long can price increases last? TrendForce data shows that Q2 NAND contract prices rose 70%-75% quarter-on-quarter, but Q3 gains have sharply dropped to about 20%.
The price hike wave has faded, so who's swimming naked?
Second, the consumer business collapsed.
The data center business is indeed strong—$2.977 billion, exceeding expectations. But the consumer segment was only $556 million, 36% less than the market expectation of $874 million, and down 32% year-on-year.
On one side, AI customers are placing orders like crazy; on the other, ordinary consumers can't afford to buy.
Two worlds, one company.
"What about long-term agreements? Isn't the $93.9 billion guaranteed income locked in? ”
That's right. Eight long-term NBM agreements cover over 50% of supply in fiscal year 2027 and about two-thirds in fiscal year 2028. Based on guaranteed prices, all long-term contracts can generate at least $93.9 billion in revenue.
But how did the market react?
Fall.
Why?
Because long-term contracts lock in volume, not price. If NAND prices are halved next year, no matter how large the guaranteed revenue figure is, actual profits will shrink significantly.
The market's concern has never been whether SanDisk can be sold.
The market is worried about whether SanDisk can still sell at this price.
That's the significance of Investor Day on August 13.
CEO David Goeckeler and CFO Luis Visoso will personally take the stage and spend several hours explaining three things:
First, HBF high-bandwidth flash memory—when exactly will it be commercialized?
On August 4th, SanDisk and SK Hynix jointly released the world's first HBF technical specification. HBF is positioned as a new storage layer between HBM and SSD, with a maximum single-chip capacity of 512GB. This is SanDisk's biggest card—if HBF becomes the standard storage layer for AI inference, SanDisk can transform from a "NAND seller" to a "core supplier of AI infrastructure."
Second, can NAND prices really hold steady?
SanDisk previously predicted that the global NAND market would exceed $300 billion by 2026 and approach $500 billion by 2027. But price increases in Q3 have already slowed significantly. Management needs to answer a tough question: Once the dividends from price hikes have been absorbed, where will the next growth point be?
Third, how will the 14 billion yuan buyback be spent?
Is it to support the market, or do they truly think the stock price is undervalued?
To be honest—
SanDisk is now in a very awkward position.
Bulls say: AI storage demand is just beginning, HBF is the next super trend, long-term contracts have locked in revenue for four years, and now the stock price has been halved from its peak—it's a golden pit.
Bears say: the β of price hikes has been eaten, the consumer business is collapsing, the NAND cycle turning point has arrived, and the current level is still too expensive.
Both groups have valid points.
But the market only recognizes one thing: on Investor Day, can management come up with something to silence the bears?
Finally, here's a framework for thinking—
If you're waiting for Investor Day, what are you waiting for?
It's not about waiting for stock price fluctuations. It's about waiting for these three signals:
Signal 1: Is HBF's commercialization timeline exceeding expectations? —If management says "mass production next year," that's a nuclear-level boost. If it's just "still exploring," the stock price will keep falling.
Signal 2: Has the long-term NAND price guidance been raised? —If management dares to say "prices will hold steady in 2027," that's confidence. If they are vague, it's guilt.
Signal Three: Are there any new super clients signed? — At the call, management said, "After signing the largest client, additional demand will be added." If investors reveal another giant name during the day, their sentiment will completely reverse.
These three signals are more important than any candlestick.
August 13, 9:00 AM Eastern Time.
Is it a lifeline, or just another stab?
We'll see when the time comes.
$SNDK $SKHY $SAMSUNG #闪迪8月13日投资者日临近, the divergence in the financial report remains to be resolved 🔥#财报观察员：AI infrastructure earnings reports take the stage one after another. This week, AI infrastructure delivered intensive results with explosive numbers, but stock prices followed a different path.
NVIDIA's Q1 revenue was 81.6 billion, with data center revenue at 75.2 billion accounting for 92%. It just partnered with Apollo, Blackstone, and four other major institutions to plan a $500 billion AI infrastructure financing. Broadcom's Q2 revenue was 22.2 billion, with AI semiconductor revenue up 143% year-over-year. AMD's revenue was 11.5 billion, with data center revenue doubling to 6.7 billion, hitting a historic high, yet its stock fell 9% after hours. Micron's Q3 revenue was 41.5 billion, up 346% year-over-year, with a gross margin reaching 84.6%, but its stock price retreated from the peak. Intel's Q2 results exceeded expectations on both fronts, clearly benefiting from AI.
The five major cloud providers are expected to spend 805 billion in capital expenditures by 2026, more than doubling from 2024. Private capital is rushing in.
But doubling revenue is no longer enough. AMD fell 9% after hours, and SanDisk dropped 12% after its earnings report — good performance is expected; the key is whether guidance can continue to exceed expectations. AI infrastructure is moving from "storytelling" to "accounting" mode.
BTC follows the Nasdaq; as long as AI hardware demand doesn't collapse and tech stock valuations don't adjust downward, BTC's downside remains relatively controllable. But Harmak just said "multiple rate hikes possible," and macro liquidity is the real ceiling.
👇 How long do you think this AI infrastructure wave can keep surging? Let's discuss in the comments.Jensen Huang appeared on CNBC live alongside Goldman Sachs, BlackRock, BlackRock, KKR, Brookfield, and Apollo to explain the new $500 billion AI financing plan:
1. Computing power is as essential as water and electricity
This is the biggest fundamental transformation in the computer industry in 60 years
This may benefit most even usable fuel generator companies with $BE, but in the long run, transformers and grid infrastructure still depend on the situation
2. Demand is too strong, you have to raise a lot of money
Not only do they need to buy chips, but they also have to compete for land, build power grids, and build factories. Just building a gigawatt-sized AI data center costs as much as $50 to $60 billion
The main issue is the mismatch between government and enterprise demand, so companies in this area can pay more attention
3. NVIDIA is no longer just a chip seller
Now it's directly upgrading to an AI factory platform. This platform is universal across industries and can run any AI model, breaking the dilemma of over-reliance on other companies in some areas. #FinancialReportObserver: AI infrastructure earnings report relay debut Today's Asian Handicap Summary (August 11)
Japan's market closure (Yama-no-Hi) has led to clear market differentiation.
Hong Kong stocks weakened: Hang Seng Index closed at 25,652 points, down 1.1%; Hang Seng Tech fell 1.93%, with overnight Nasdaq pulling back drag down the tech sector, with NIO and BYD Electronics leading the declines. Energy stocks rose against the trend, CNOOC led the gains, and Brent approached 85 for a transmission effect.
A-shares weak in Shanghai with strong Shenzhen stocks: The Shanghai Composite ended six consecutive gains, while the ChiNext Index opened lower but rose 1.41%. The style shifted from high-level tech to consumer and pharmaceutical sectors, with over 3,100 stocks falling. Northbound + leveraged funds saw a net outflow of about 12 billion yuan, showing a clear pattern of stock competition.
Other markets: South Korea and Australia followed with slight declines; the Asian session was generally suppressed by both oil price transmission and technology synergy, showing a pattern of strong energy and weak technology.
Tonight's U.S. stock market outlook
Tonight's focus: 6 points NFIB SME Confidence Index, 10 points for existing home sales. The real main theme of the market is Wednesday's CPI, with funds leaning defensively before that.
Pre-market futures were weak, with key S&P levels: 774.74 above is confirmed bullish, while a drop below 771.60 signals weakness. Tech giants remained resilient last week, but the continued weakness in semiconductors (Philadelphia Semiconductor Index fell over 2% yesterday) poses a hidden risk. High oil prices and a 10-year yield of 4.70% weigh on growth stock valuations.
Overall judgment: Before CPI is implemented, narrow fluctuations are unlikely to break the deadlock, and the energy sector may continue to benefit in the short term from sentiment in Hormuz.Most people say that staking 42 million ETH is a supply advantage, but I don't see it that way
The staked amount has surpassed 42 million, accounting for nearly 35% of the total supply. Many accounts are saying: ETH is becoming increasingly scarce, supply is locked, and price pressure is easing.
I understand this logic, and on the surface, it does hold true. But if you look closely at what happened to Ethereum this year, you'll find that behind the staking rate surge lies a more complex, even somewhat ironic, story.
On August 4, Justin Drake, together with five other Ethereum Foundation researchers, submitted EIP-8361, with the core mechanism being "Tapered Issuance Burn": as the staking ratio increases, the burned proportion of validator rewards also rises until the staked amount reaches 50% of the total supply (about 60.25 million ETH), at which point the new consensus layer issuance drops to zero.
This proposal made me think for a long time.
On the surface, EIP-8361 is addressing the problem of "centralization caused by excessive staking"—yes, when 35% of ETH lies in staking contracts, plus Lido alone accounts for over 30% of validators, the risk of centralization is real.
But here's the question: who is most unfavorable to this proposal?
The biggest disadvantage is for those currently staking. If you stake ETH today, you're getting about 3.5%-4% annualized returns. Once EIP-8361 is passed and the staking rate continues to rise, your earnings will be automatically diluted by the system until they reach the 50% threshold. In other words, more and more people are staking, pushing for a tipping point where their returns are diminishing faster.
It's a bit like the story of everyone desperately pouring water into the pool, unaware that the hole in the pool is getting bigger and bigger.
Now, let's talk about the DeFi side. The base yield of Ethereum staking has long been regarded as the DeFi world's "zero-risk rate anchor"—lending protocols like Aave and Compound set rates based on this anchor to some extent. If EIP-8361 lowers or even zeros this underlying yield, liquid staking projects (Lido, Rocket Pool) and LRT protocols (EigenLayer's EIGEN staking logic) that rely on Ethereum staking yields as product narratives will be hit by valuation shocks.
The community controversy over EIP-8361 is still significant. Some voices on the Ethereum Magicians forum bluntly say: the proposal was submitted before the Hegotá upgrade deadline, leaving very little time for community discussion. This is not an ordinary parameter adjustment; it is a fundamental change in Ethereum's monetary policy, yet it is being pushed in like an emergency bill. This procedural issue itself deserves special attention.
My current judgment is: the probability of EIP-8361 being implemented in 2026 is low, but its very existence has already cast doubt on ETH's "monetary expectations"—what exactly will Ethereum's future issuance policy be? Who decides that? This uncertainty creates friction for institutions allocating ETH.
This is my current understanding, but I leave myself 30% room for reversal, because if EIP-8361 passes as a modified version with a sufficiently long transition period (the proposal mentions 18 months), the impact may be milder than I expected. For those interested in ETH staking mechanisms, do you think the impact on DeFi will be more severe than the consensus layer issuance itself, or will it be less significant?
#现货ETF资金分化, BTC selling pressure remains 🍎 APPLE × CHANGXIN: THIS IS BIGGER THAN A SUPPLIER STORY
Apple reportedly testing ChangXin Memory Technologies’ DRAM chips is getting attention for a reason.
The bigger signal isn’t simply “Apple wants another supplier.”
It’s that the global memory market is becoming so tight that even a company with Apple’s purchasing power is looking for additional sources of supply.
📌 Why ChangXin matters
ChangXin has been expanding its position in the global DRAM market and is increasingly being viewed as a serious fourth player alongside Samsung, SK Hynix and Micron.
But there’s an important detail:
ChangXin reportedly isn’t trying to win customers by simply offering the cheapest chips. With domestic demand already absorbing much of its capacity, its pricing power appears stronger than many expected.
That changes the narrative.
This isn’t just about replacing one supplier.
It’s about China building a more competitive position in a strategically important semiconductor industry.
🔥 And what about BTC?
There’s no direct Apple → ChangXin → Bitcoin connection.
The potential link is macro:
AI infrastructure is consuming enormous amounts of computing hardware → memory demand stays elevated → semiconductor prices remain under pressure → companies raise product prices → inflation expectations can stay sticky.
Short term, persistent inflation can keep pressure on risk assets.
But over the longer horizon, continued monetary and technological expansion strengthens the debate around scarce, non-sovereign assets.
And that’s where $BTC becomes interesting.
The real takeaway?
Apple testing ChangXin may be less important than what it says about the memory market itself.
When supply becomes scarce enough that even Apple starts looking for alternatives, the semiconductor cycle deserves attention.
And where capital expenditure, AI infrastructure and inflation go…
BTC eventually gets pulled into the conversation. 👀
$BTC $ETH $BICO
#Bitcoin #BTC #Ethereum #Apple #Semiconductor #AI #Crypto #Orbit Altcoins: A survival game with a life-or-death experience
If you're still fantasizing about getting rich overnight with altcoins, take a look at this data: tracking 1,972 tokens with market caps exceeding $50 million, only 4.1% outperformed Bitcoin, with a median return of 97% and 73% of tokens losing over 90%. This is not a coincidence, but the fate of the altcoins.
Why are altcoins doomed to zero?
The issuance threshold is extremely low, with thousands of new projects flooding in every year, diverting limited funds; Most projects lack real commercial value, with valuations supported entirely by narrative and sentiment; The token unlock mechanism is a ticking time bomb—institutions enter at extremely low costs, and retail investors take over and face massive selling pressure. Even more brutal, when the market panics, liquidity instantly evaporates and you don't even have a chance to cut your losses.
Survival rule: If you insist on participating,
First, reserve over 90% of your position in Bitcoin, and use altcoins only as "lottery positions"—don't mind losing everything; Second, only select leading projects in the sector—the top two in public chains, DeFi, AI, etc. ($SOL, $PUMP, $UNI, $HYPE, $ETH); the rest are basically cannon fodder; Third, never chase highs, only test small positions when the market stabilizes and sector rotation is early; Fourth, strictly set stop-loss orders, exit unconditionally at -20%, and never add positions to dilute costs.
Most altcoins will ultimately become your investment journey. Remember: in this market, survival is more important than anything. Slow is the fastest path.
$BTC $DOGE (1H) – Ascending Moving Average Ribbon
Bias: LONG
Entry Zone: 0.07005 – 0.07020
Stop Loss: 0.06975
TP1: 0.07050
TP2: 0.07090
TP3: 0.07130
Why this setup:
Price maintains a clean bullish curve off the 0.06940 base, holding support above the stacked MA5 (0.07009), MA10 (0.07004), and MA20 (0.06989) levels to target a breakout of 0.07036.
NFA – Educational purposes only.
#AIInfraEarningsWatch #CPIToResetFedBets #AIInfraFundingDiverges $DOS 这波到顶了吗？
dos刚刚那根针再次突破了新高0.5391，这会肯定有很多宝子在想这个位置还能不能追涨！
从当前的盘面来看新币上线波动肯定大，也会有大波抛压，刚刚还上线了泡菜国交易所利好已经兑现，这个位置还是不要盲目的追涨。
杨哥在0.485附近进场的多单已经浮盈了65%，已经带了成本损格局一下
$BEAT $BICO #财报观察员：AI基建财报接力登场
#本周三CPI公布，9月加息定价会改写吗？ #现货ETF资金分化, BTC selling pressure still Spot BTC ETF saw a net outflow of about $144.6 million on August 10, and ETH ETF also turned to a net outflow of about $14.6 million. Meanwhile, on-chain selling pressure has not disappeared. Lookonchain monitored that one whale sold a total of 7,513 BTC over the past three weeks, worth about $487 million; Another suspected miner-related address transferred 6,494 BTC to Binance over the past three weeks, about $420 million.
Interestingly, with such large chips being pressed into the market, BTC is still holding near $64,000.
So now, I am no longer simply bearish.
What is truly worth watching is: if ETF funds turn positive again and whale selling pressure gradually weakens, the $64,000 area could be the spot for bulls and bears to reprice; But if ETFs continue to flow out and on-chain coins keep being sent to exchanges, the support below will be further tested.
The next biggest variable is the CPI. The U.S. Bureau of Labor Statistics confirmed that the July CPI will be released at 8:30 AM ET on August 12.
This round isn't without buyers, but rather they're head-to-head with whale selling pressure. Whoever exhausts first will see the next direction basically emerge. @OKX planet Traditional assets such as US stocks are being implemented through the integration of wallets and on-chain protocols, finding a convergence point for cross-market liquidity on-chain. Against the backdrop of high-interest US stocks and crypto assets linking to crypto assets, asset listing brings direct ecosystem use cases and value capture mechanisms to $OKB. If traditional financial assets such as bonds and funds continue to expand, the momentum for token valuation reshaping will be released simultaneously. If cross-jurisdictional regulatory constraints tighten or on-chain US stock liquidity is under pressure, the upward pace will slow accordingly. The next step is to focus on on-chain US stock trading volume and premium changes.
#贝莱德IBIT换购门槛降至100万美元 #CLARITY表决推迟至9月, the supervision window has been moved backwardCrowding and Crowding List
High fees are not the conclusion, and low rates are not opportunities; what really matters is the return on your position.
$DOS Current rate -0.7849%, closed in the past 24 hours -2.344%, at the 20th percentile of the most recent sample. Price decline increases positions, new leveraged funds are participating in this downward phase. Bears continue to expand positions at high costs; this is not a bottom-fishing signal; the real risk is adding positions without falling. Historical samples have only 5 settlement points; percentiles are temporarily auxiliary.
$SKHYNIX Current rate +0.0996%, closed in the past 24 hours +0.167%, at the 75th percentile of the most recent sample. Prices are down, so are open positions; the ebb tide of position is more certain than direction attribution. Open interest is declining, crowded positions are retreating first, and the current focus is when the reduction pace will slow down.
$BEAT Current rate +0.0598%, closed in the past 24 hours +0.222%, at the 97th percentile of the most recent sample. Expanding positions while falling is accompanied by selling pressure with new positions, but still cannot rely solely on OI to confirm the direction of short positions. Prices fall, OI increases, while rates remain positive; this misalignment is more sensitive to bulls.AI Is Bidding For The Rails
Compute Eats Capital
The AI trade moved further out of the demo room and into project finance: Nvidia is reportedly lining up a 500b financing effort with Apollo, Blackstone, BlackRock, Goldman Sachs, KKR and Brookfield, while Big Tech AI CapEx is estimated to hit a record 2.4% of US GDP in 2026. BlackRock’s Larry Fink said AI will require more than 70 gigawatts of power, and Meta announced a 1b fund for US cities hosting its AI data centers. The non-obvious crypto angle is power competition: Keel shuttered all US bitcoin mining operations to pivot fully to AI, which says the compute stack is now bidding directly against hashpower for infrastructure.
Agents Hit Reality
The cleanest AI risk signal today was not a lab benchmark; it was an agent booking a gym class, reportedly finding a software vulnerability and kicking another member off the waitlist to secure a spot. Decrypt also flagged hidden text in PDFs hijacking an AI assistant, while North Korea’s Kimsuky is reportedly integrating AI into cyberattacks targeting crypto and finance. That is the uncomfortable middle ground: agents do not need to be superintelligent to become operational security problems; they only need permissions, tools and sloppy boundaries.
$BTC
#BTCETHETFFlowsDiverge I'm bearish on SpaceX
Starting August 20, internal shares will be unlocked in batches. In September, 44% of internal shares can be sold, with pressure continuing until December 8. By then, the free market will expand nearly tenfold, greatly increasing the number of chips. Refer to Facebook's 2012 account on August 20. When the negative news materializes, it will turn positive, so go long either way, referencing the first unlocked rally on August 6
$SPCX Darcy, co-founder of Flash Rescue, revealed that the ODY fund project targeting Chinese victims issued an additional 10 billion tokens on July 28 and withdrew about 15 million USDT from the trading pool. Currently, the amount of ODY-related funds collected by the company is at least of a relevant scale.Performance soared 372%, but the stock price fell 47%—SanDisk's Investor Day: Is it a "self-rescue" or a "reversal"?
Have you ever seen a company like this?
Revenue reached $8.97 billion, a year-on-year surge of 372%.
Profits are 135 times higher than the same period last year.
The gross margin was 84.6%, an absurdly high level.
The board also approved a $14 billion buyback plan.
Then, the stock price dropped more than 10% over two days.
From a historic high of $2,354 in June, it fell all the way down to $1,238—a 47% evaporation of market value, losing over $150 billion.
You read that right.
This is SanDisk.
On the night the financial report came out, I stared at the screen for ten minutes.
Revenue exceeded expectations, profit exceeded expectations, gross margin exceeded expectations, and buybacks exceeded expectations—four "exceeding expectations" stacked together, causing a 7% drop in after-hours trading.
What logic?
Because the guidance for the next quarter is "not impressive enough."
SanDisk expects revenue for the next fiscal quarter to be between $10.3 billion and $10.8 billion, with a median of $10.55 billion—while Wall Street's most optimistic expectation is $11.16 billion.
Just 600 million short.
A 600 million yuan gap, the market punishes it with a market value of 150 billion yuan.
Wall Street isn't just about "goodness" now. It's about "perfection."
What's even more heartbreaking is the truth behind SanDisk's recent price increase.
Many people think the 372% surge in revenue is due to a global surge in storage demand.
Wrong.
Management personally said: Of the 51% quarter-on-quarter revenue growth, only one-third came from increased shipments—the remaining two-thirds were all from price hikes.
This is not a boom supported by demand.
This is a bubble inflated by price hikes.
TrendForce data shows that in the second quarter of 2026, NAND contract prices rose 70% to 75% quarter-on-quarter. But in the third quarter, the increase plummeted to around 20%.
The wheels of price increases are slowing down.
But the other side of the story is also quite interesting.
SanDisk has signed 10 long-term "new business model" agreements, securing supply for eight core customers over the next four years.
More than half of the supply for fiscal year 2027 has already been locked in early, and about two-thirds for fiscal year 2028 are already arranged.
These agreements are expected to generate minimum revenues of $93.9 billion and customer default protections of $16.5 billion.
The CEO said something during the conference call that left a deep impression on me:
"In the past, we could only predict demand within three months, but now we hold over four years of lock-in procurement volume."
From "three months" to "four years"—this is the true qualitative change.
Therefore, on the August 13th Investor Day, SanDisk's management will answer only one core question:
Are you a cyclical company that makes a living by price hikes, or a platform company that survives cycles through long-term agreements?
The market now chooses to trust the former—so the stock price has been halved from its peak.
But if management can prove the latter at Investor Day—proving that the NBM protocol is not just for show, to prove that AI storage demand is not a short-term impulse but a long-term trend, and to prove that the 84.6% gross margin is not the peak but the new normal—
Then the current $1,238 could be the future bottom.
The market never fears companies making less profit.
What the market fears is—you don't know if you can still make this much next year.
SanDisk attempts to answer this question with 10 long-term contracts and a minimum income guarantee of $93.9 billion.
But investors have not yet been convinced.
On August 13, it all depends on whether the management can tell this story well.
$SNDK $SKHYNIX $SAMSUNG #闪迪8月13日投资者日临近, divergences in the financial report remain to be resolved Broadridge disclosed that its distributed ledger buyback platform DLR processed over $8 trillion in transactions in July, averaging about $365 billion per day, a year-on-year increase of 28%.
This figure is not the TVL of public blockchains, nor the $8 trillion inflow into crypto. It corresponds to institutions using distributed ledgers to complete buyback settlements and real-time transfers of tokenized collateral within existing trading and clearing systems.
Such news does not stimulate the price but indicates that on-chain settlement on the institutional side has already generated considerable business volume.AI infrastructure is becoming as much a financing story as a technology story. Nvidia’s platform with BlackRock, Blackstone and Goldman Sachs targets more than $500B in third-party capital for customer data centers and GPUs, though deals remain pending. Intel, meanwhile, may lift its own offering from $15B to about $20B after attracting over $100B in orders.
The distinction matters: external capital can support customer demand, while equity issuance funds Intel’s own buildout and raises dilution questions. As spending scales, funding structure may become a sharper valuation signal than headline demand alone. Not advice, just analysis.
#AIInfraFundingDivergesMarkets Are No Longer Watching the Deal. They're Watching Whether It Can Be Implemented.
Markets initially welcomed reports that Iran and Oman had reached a preliminary understanding on new shipping arrangements through the Strait of Hormuz.
Now, attention has shifted.
The biggest questions are no longer about whether an agreement exists—but whether commercial shipping can actually resume under increasingly complex conditions.
Reports suggest unresolved issues include sanctions compliance, insurance coverage, transit rules and enforcement responsibilities. Iran's parliament is also considering tighter transit regulations, adding another layer of uncertainty.
This matters far beyond the energy market.
The Strait of Hormuz handles roughly a fifth of global oil shipments, making it one of the world's most strategically important trade routes.
If disruptions persist, higher oil prices could keep inflation elevated, complicate central bank policy and delay expectations for easier monetary conditions.
That's where crypto becomes part of the story.
Bitcoin doesn't trade on shipping lanes.
It trades on liquidity.
And liquidity is heavily influenced by inflation, interest rates and broader macro conditions.
Sometimes the most important crypto catalysts begin far outside the crypto industry itself.
Do you think geopolitics will play a larger role in crypto markets over the coming years than many investors currently expect?
Share your thoughts below 👇 #HormuzDealUnresolved Builder fireplace, affiliated with Polymarket Developers, announced it will be shutting down
From the mainnet launch in late January to its closure in just over half a year, it raised $1.5 million during that period
After the World Cup, the prediction market may be entering a winter, which will be a blow to smaller terminal platforms
Polymarket Builders have over 450+ units, but weekly trading volume has declined for five consecutive weeks, and many smaller platforms may have shut down one after anotherSpaceX's short squeeze this week was quite resolute, not even looking back three times, allowing the bulls who had been suppressed for two months to breathe a sigh of relief.
Yesterday, it closed up 4.23% at 138.74, maintaining stability. Besides itself, the market speculates that RKLB's earnings report and the failed launch of Zhongxing 4B may also affect SPCX's stock price, but in reality, the main logic is its own independence.
For commercial space, launch itself is just infrastructure, while the functions of satellites are the business itself. Now the entire industry is turning into a mini SpaceX; the battle over business models is over, and what remains is the battle of scale.
The failure of the Long March 7 modification will raise SpaceX's reliability premium and put the brakes on China's narrative of weakening SpaceX's technological scarcity. It may help the current short-term squeeze, but it won't be decisive. If the final fault investigation confirms the problem is with the YF-100 engine, the shared heart of the new generation Long March family, then it's very likely that the bear catalyst hanging over SPCX's Q4 will be delayed.
Tomorrow night's CPI data should not be lower than last month's, since oil prices have already started rebounding since early July. But it shouldn't be very high, because oil prices have a lag in transmission and most of the rebound will be reflected in next month's data. Judging from today's gold price movement, the current pullback feels more like a rally to take profits and wait for CPI, rather than proof that the market has completely ruled out the risk of a rate hike. Therefore, it's not advisable to be bullish in US stocks before the CPI data is released $SPCX 🚨 On August 14, the SEC initiated its first major crypto rulemaking
The U.S. SEC will hold a public meeting this Friday to formally propose "Reg Crypto," establishing a tailored issuance mechanism for specific crypto asset investment contracts. This is the SEC's first formal rulemaking for crypto initiation.
The highly anticipated CLARITY Bill was delayed in the Senate and failed to advance before the recess. However, the SEC's decision to accelerate action shows that regulation has not stalled due to congressional blockages.
Market impact: There may be short-term volatility, but in the medium to long term, it will provide a clearer compliance pathway, helping to attract institutional capital and reduce uncertainty.
My view is—the market is no longer betting solely on clear legislation. With the SEC proactively pushing rulemaking, crypto market prices should become more stable, no longer overly reliant on single legislative outcomes. Regulatory clarity is increasing, which is good for long-term development.That's ruthless! Nvidia's "getting something for nothing" tactic, crushing Intel to the ground? 🤯
Truly amazed by NVIDIA's financial skills! 👇
🔥 Nvidia's Actions:
They teamed up with BlackRock and Blackstone to form a $500 billion financing platform.
To put it simply: I want to sell a shovel, but the client can't afford it? No worries, I borrow money from a Wall Street tycoon to the client, and the client will use the money to buy my shovel!
This is a top-tier business closed loop: not only do they sell goods, but they also make money from finance, barely losing money.
🩸 Intel's situation:
Looking at Intel, it raised 20 billion yuan by issuing new shares and diluted its equity. Although subscriptions were booming (over 100 billion in demand), this was clearly "surviving from blood loss," forcing them to seek money from the market to make chips.
💡 Insights for retail investors:
When looking at a company, it's not just about its products, but also about its 'ability to make money.' Nvidia, which can mobilize external capital to help run its business, is the true king.
Beware of "all the good news is exhausted." Although Intel's replacements are positive (with funds to expand production), the short-term equity dilution is real, so don't rush blindly.
AI infrastructure is still in the money-burning phase. Whoever can get the money at lower cost will survive until the end. So far, Old Huang has hit the jackpot.
Do you think Intel can turn things around this time?
#AI基建融资升温, Nvidia and Intel are diverging in their paths 1. First News: Trump Reveals Three U.S. Strategies Regarding Iran
1. Core information breakdown
- Source: Xinhua News Agency cited a report from Qatar's Al Jazeera on the 11th, stating Trump's public statements in an interview with "Real American Voice."
- Core content: The U.S. has three strategies toward Iran: monitoring the deterioration of Iran's situation, launching a fierce strike on Iran, and putting pressure on Iran's economy; At the same time, it is clear that the U.S. controls a large amount of Iranian funds and assets, completely under U.S. control
- Market sentiment tags: 6 positive, 20 negative; overall market interpretation is biased bearish
2. Analysis of geopolitical and economic impacts
- Geopolitical risks are rising sharply: Among the three strategies, the military option of "launching a fierce strike on Iran" directly shatters expectations of relative stability in the Middle East and may trigger military conflict risks in the region. As a major global energy exporter, Iran's conflict would directly impact the global crude oil supply chain, driving up oil prices and global inflation expectations.
- Iran's economic pressure intensifies: The U.S. already controls a large amount of Iran's financial assets, and combined with its economic pressure strategy, Iran's foreign trade and financial activities will be further restricted, increasing domestic inflation and exchange rate collapse risks, and also affecting the supply stability of the global energy market.
- Rising global risk aversion: Geopolitical conflict risks directly reduce risk appetite in global markets, causing funds to flow into safe-haven assets such as gold, US Treasuries, and the yen, putting pressure on global stock markets—especially high-risk emerging market equities.
2. Second News: Bitunix analysts interpret the impact of nonfarm payroll data and exchange rate interventions
1. Core information breakdown
- Core Event: U.S. nonfarm payrolls unexpectedly fell by 23,000 in July, and combined with Japan-U.S. exchange rate interventions, global concerns about high funding costs have intensified
- Market core focus: US July CPI data and capital efficiency in the AI industry
- Core interpretation logic: The decline in nonfarm payroll data → cooling the U.S. labor market→ shattering expectations of economic overheating→ the market reassessing the Fed's monetary policy path→ increasing the urgency of rate cuts→ but economic weakness weakens growth support→ and risk aversion further increases funding cost pressures→ data becomes a key turning point to reverse, pointing to a substantial weakening of U.S. economic growth momentum
- Market sentiment tags: Positive 3, Negative 4; the overall market interpretation is biased toward negative sentiment
2. Macro and market impact analysis
- Federal Reserve monetary policy expectations have completely reversed: Nonfarm payrolls are a core leading indicator of the US economy. The unexpected decline directly proves the cooling of the US economy. Market expectations for the Fed will shift from "maintaining high rates longer" to "cutting rates earlier/faster," putting pressure on the US dollar index and causing Treasury yields to fall in line with rate cut forecasts.
- The contradiction between funding costs and asset valuation: Although rising rate cut expectations will keep funding costs down in the long term, short-term market risk aversion to economic recession can actually push up actual funding costs, putting pressure on valuations of global high-risk assets (especially high-valuation growth stocks and AI sector assets), because rising funding costs directly compress asset valuation space.
- Intensified exchange rate market volatility: Japan-US exchange rate intervention means Japanese authorities have begun intervening in yen depreciation. Yen appreciation will directly affect the profits of Japanese export companies and also change the flow of funds in global exchange markets. The USD/JPY movement will become a key variable affecting global stock and bond markets.
- Core market anchor for the future: July CPI data is a key constraint on the Fed's monetary policy—if inflation remains high, the Fed's rate cut pace will be limited; If inflation falls in tandem, rate cut expectations will further strengthen. At the same time, the capital efficiency of the AI industry determines whether the previously favored AI sector can maintain high growth, directly affecting the performance of global tech stocks.
3. Summary of the overall market impact of the two news stories
Both pieces of news point to a decline in global market risk appetite and a rise in risk-averse sentiment:
1. Geopolitical Level: The U.S. tough stance on Iran has directly increased Middle East geopolitical risks, benefiting safe-haven assets like crude oil and gold, and negatively affecting global risk assets.
2. Macro Perspective: U.S. nonfarm payroll data was unexpectedly weak, indicating weakening economic growth momentum. Although expectations for rate cuts have risen, concerns about a short-term recession will weigh on risk assets, while dollar pressure will benefit non-U.S. currencies and emerging market assets.
3. Overall Transmission: The combination of these two pieces of news will drive global capital to shift from high-risk stock and growth stocks to safe-haven assets such as gold, US Treasuries, and the Japanese yen, leading to significant short-term market volatility.
$BTC $ETH
#现货ETF资金分化, BTC selling pressure remains A Wall Street Legend from a Century Ago: How Did Jesse Livermore Foresee the Crash?
Jesse Livermore is one of the most legendary speculators of the 20th century, known as the "Wall Street Bear." His two most famous wins are:
The 1907 financial panic shorted the market
Before the Wall Street crash in 1929, huge short positions were established
One of the 1929 trades reportedly earned him about $100 million (the amount at the time), making it one of the most famous short trades in financial history.
But what truly deserves research is not "he guessed the crash," but rather:
He is not predicting the date, but observing that the market structure is collapsing.
1. He looks at the 'overall market,' not individual stocks
In his early years, Lemore liked to study individual stocks, but later he changed:
Buy long in a bull market, short in a bear market, and follow the main market trends.
He believes that stocks are not independent movements but are controlled by market trends.
This idea and the present are:
Dow Theory
Market breadth
Cash flow
Macro cycle
Actually, it's very close.
2. Before 1929, he saw five warning signs
(1) Everyone believes 'this time is different.'
The 1920s U.S. stock market bull market:
Retail investors entered in large numbers
Leveraged stock buying is prevalent
Stocks have become a tool for everyone to get rich
When the market begins to emerge:
"Stocks only go up"
This was a very vigilant signal for Limoire.
Today is similar:
AI cannot be bubbled
Tech stocks are always reasonably overvalued
Every pullback is a buying opportunity
(2) Trading volume expands, but price advancement weakens
Lemore studies "price behavior."
He will observe:
Is it easy to rise?
Is the decline starting to accelerate?
Is buying unable to push prices higher?
For example:
Stock price:
100 → 120 → 130
However:
Trading volume increased
The gains have narrowed
Representative:
Large funds may be being distributed.
This is the same as now:
Wyckoff Distribution
Volume Spread Analysis
Smart Money Concept
Very close.
(3) Leading stocks are starting to lose momentum
At the end of a bull market, typically:
Phase One:
Quality stocks led the gains
Phase Two:
second-tier stocks caught up in the gains
Stage Three:
Junk stocks are soaring
The market before 1929 was similar.
When weak companies start to surge, it means the last bit of liquidity in the market is burning away.
(4) Excessive credit leverage
Before 1929, many investors used margin to buy stocks.
Market Uptrend:
Leverage → more buying → higher prices
Foam forms.
But on the other hand:
Prices fall → margin calls → forced to sell → crash.
This is also today:
Financing transactions
Futures leverage
Cryptocurrency liquidation
The same logic.
(5) He waits for 'market confirmation'
Lemoore doesn't short-sell immediately at the sight of a bubble.
Its core:
The market proved me right, so I increased my position.
For example:
Let's take a look:
Key support was broken
The rebound is weak
The trend has turned bearish
Then gradually increase the shorts.
This is very important.
Many traders:
Seeing overvalued → immediately becomes empty
Results:
The bubble can last for half a year or even several years.
3. Lemoire's 1929 operating model
General process:
Phase One:
Hold long positions in a bull market.
↓
Phase Two:
Starting to feel the market is abnormal:
Stock valuations are extreme
Speculative mania
The trend is starting to weaken
↓
Stage Three:
Establish a short position.
↓
Phase Four:
After the crash is confirmed, increase your position.
During the 1929 crash, he made huge profits.
4. But Lemoore's biggest lesson: Being able to predict doesn't mean you can keep your money
His tragedy:
He once made a fortune
Later, he went bankrupt multiple times
In the end, his life ended in tragedy
Causes:
It's not that the technique is bad.
Instead:
There is no permanent control over risk.
For example:
Overconcentration
Overleverage
Trust your own judgment
5. If applied to today's market,
Lemore's method can be translated as:
Bull Market End-of-Market Checklist
✅ Market valuations are extreme
✅ Retail investors are frenzied
✅ All the media are bullish
✅ Weak stocks surged
✅ Trading volume has increased, but the rate of increase has declined
✅ The leading stock fell below key moving averages
✅ Credit leverage increased
When multiple occurrences occur:
Not shorting immediately.
Instead:
Reduce positions and wait for market confirmation.
Actually, the capital rotation in the latter half of the AI bubble that you studied before—Wyckoff, SMC, Willy Mid Tri + SMA200—is very similar to Leemore's thinking:
He wasn't looking at the "news," but:
Capital→ Trends→ Crowd Psychology → Market Structure.
Lemoire's most iconic quote:
"The market never makes mistakes; only people's opinions can be wrong."
This is also why methods from 100 years ago are still being studied today.#USWeighsIranStrike #HormuzDealStillPending
◇ Beijing's role in peace agreements in the Gulf!
* Iran's longest-standing and largest oil buyer
* 40% of China's oil imports come from Hormuz
* As part of a military partnership for Iran
* There is a certain influence in the Bay Area
* Successfully mediated the Saudi Arabia-Iran conflict in 2003
* Taking advantage of the conflict to reduce the prestige of the United States, strengthen the image of the new empire with the slogan "Responsible, no military intervention"
◇ What can Beijing do by participating in the mediation process?
* As its largest military customer/partner, Beijing can fully coerce or offer preferential packages of economic/military assistance to Tehran to reduce tensions and reopen the Strait of Hormuz.
* In fact, China does not want the United States to win, and even more so does not want Iran to continue to escalate tensions, causing the Maritime Route of Hormuz to be blocked. Direct impact on China's energy security.
* Maintain constant contact with Washington as Tehran refuses to negotiate directly.
◇ Results and time to reach a peace agreement?
* Shortly after Beijing announced that it would directly participate in the peace agreements, attacks by both sides were on pause.
* The Gulf states (Qatar, UAE, Saudi, Oman, etc.) are starting to get tired of the protracted war, which directly affects the economy and civilians here.
* In the immediate future, there will be an interim agreement (with time) to open the Strait of Hormuz, reduce or suspend military operations. A nuclear deal will be the final step towards completing the peace mission for the parties.
* It is highly likely that there will be an agreement this week to open the gates of Hormuz, and by the end of August there will be commitments to stop military activities between the parties (including Lebanon and Palestine). Sometime between mid or late September, a nuclear deal will end the conflict in the region.
♡ The market will continue to move sideways at least until the end of September 2026 "Money in the market is rotating, but BTC hasn't taken the baton yet"
Tuesday, August 11, 2026
Q3 · Issue 98
Aspirin · Cycle analysis from a data scientist's perspective
Semiconductors have stepped down from the main stage, while healthcare, gold, and finance have moved up; BTC is still standing at the door.
At the close of the US stock market on August 10, semiconductor ETF SMH fell 2.32%, Nasdaq ETF QQQ dropped 0.31%. On the same trading day, healthcare XLV rose 1.67%, gold GLD increased 1.03%, financial XLF went up 0.40%, and the equal-weighted S&P RSP slightly rose 0.07%.
One night is not enough to declare the AI trade is fading, nor to see that a broad bull market has spread. A more market-close explanation is: the most crowded positions are starting to loosen, and funds are shifting to seek cash flow, low volatility, and safe-haven attributes.
This is very important for BTC. Every dollar sold from chip stocks does not automatically flow into the crypto market. Whether there is new money in the market determines if this is a healthy rotation or a risk reduction in a high-level market.
1. Last night's market gave a seating chart
If risk appetite is truly retreating broadly, equal-weighted indexes, finance, and healthcare usually would not all be supported simultaneously. RSP closing higher indicates buying has not disappeared; SMH's decline being significantly larger than QQQ's suggests funds are starting to be selective about the most expensive and crowded parts of the AI chain.
Long-term demand for AI has not been overturned by a single day's drop. The problem lies in price and position: when a sector already carries the most optimistic profit expectations, any slight disappointment in earnings, financing costs, or capital expenditures will cause marginal funds to look for cheaper seats. Healthcare has stable cash flow, finance benefits from high interest rates, and gold provides insurance against inflation and policy errors. Their strength last night precisely outlines current fund preferences.
Therefore, the index remaining high does not mean the internal structure hasn't changed. Focusing only on the S&P or Nasdaq's rise and fall easily misses the moment when seats have changed hands.
2. Market leverage is very high, and the index's strength increasingly depends on internal relay
The latest disclosed FINRA margin debt is $1.502 trillion, up 49.02% year-over-year. Meanwhile, the VIX remains near 15, and ICE BofA high-yield bond spreads were only 270 basis points as of August 7.
This data is awkward: positions and leverage are high, options protection is still cheap, and the credit market has not sounded alarms. The market has not yet been forced to deleverage but already lacks a thick cushion.
In this environment, the same funds find it difficult to simultaneously support semiconductors, gold, healthcare, finance, and BTC. They constantly compare odds: where expectations are too full, they withdraw a bit; where prices have fallen longer and cash flow is steadier, they sit a while. Rotation itself is not pessimistic but reminds us that the index's strength increasingly relies on internal relay.
Broad diffusion requires looser financial conditions. The equal-weighted index strengthening continuously, credit spreads remaining calm, the dollar and long-term yields falling, plus crypto ETFs flowing back in, would mark the market moving from seat swapping to adding more seats.
3. Why BTC hasn't taken the baton yet
Last week, US spot BTC ETFs had a cumulative net inflow of $865.3 million, ETH ETFs net inflow of $243.7 million. Yet prices did not form an effective breakout; BTC is still patiently hovering around $64,000.
By August 10, BTC ETFs turned to a net outflow of $144.6 million, ETH ETFs also net outflow of $14.6 million. A single day outflow does not equal a trend reversal, but it shows marginal buying is not continuous. A weekend bullish candle can come from thin liquidity; continuous ETF subscriptions are closer to real money passing the baton.
When funds flow from chips to healthcare and gold, BTC may continue to consolidate; only when new risk budgets appear, or the dollar and long-term rates fall simultaneously, will BTC more easily shift from "spectating rotation" to "participating in diffusion."
I prefer to treat ETF fund flows as a thermometer, not to treat every candlestick as the answer.
4. Tomorrow night’s CPI will separate rotation from retreat
The US CPI released at 20:30 on August 12 is the first stress test for this judgment. The data itself accounts for half; the other half is how the dollar, long-term US bonds, and market breadth vote.
If CPI is moderate, long-term yields and the dollar fall, RSP continues to outperform QQQ, and spot ETFs resume net inflows, this rotation has a chance to become a healthier upward diffusion. BTC standing back above $65,500 and holding would be a direct signal that the crypto market is taking the baton.
If CPI is hot, yields rise, SMH and BTC continue to be pressured, and even RSP, finance, and healthcare turn from gains to losses, then liquidity is declining. VIX breaking above 25 and high-yield bond spreads moving above 350 basis points would require rewriting "seat swapping" into broader risk reduction.
After CPI, four things will decide whether this judgment can be retained:
- Whether RSP can continuously outperform QQQ without the S&P breaking its upward structure;
- Whether VIX stays below 20 and high-yield bond spreads hold within 300 basis points;
- Whether BTC ETFs resume net inflows and prices reclaim $65,500;
- After CPI, whether gold's strength accompanies a dollar decline or comes from risk insurance demand.
Semiconductors only fell for one day; it's too early to write the ending for AI trading now; healthcare and gold took over for one night, but it's not enough to become the new main theme. At least wait two trading days after CPI to see if funds continue to diffuse.
The market is not short of stories; it lacks money that can support all stories simultaneously.
I will continue to record the same cross-asset risk signals, fund flows, and subsequent validations in the Aspirin · Cycle Lab group chat. Whether the judgment is right or wrong, the original record is kept for the next data to decide.
#本周三CPI公布，9月加息定价会改写吗？ Short gold!
Gold is an interest-free asset, with no interest, cash flow, or dividends. The only core anchor for pricing is the US 10-year TIPS real interest rate, which is strictly negatively correlated.
Currently, the 10-year TIPS has a real yield as high as 2.41%, with a breakeven inflation rate of only 2.26%, meaning the real interest rate has already exceeded inflation.
A simple analogy: gold = hens that don't lay eggs; interest-bearing bonds = hens that produce eggs steadily every day.
Currently, egg production yields are very high, and people are selling chickens that don't lay eggs, naturally causing prices to drop.#AI基建融资升温, Nvidia and Intel are diverging in their paths
Can SanDisk really stand atop the AI wave, or will it be unable to escape the cycle's "peaks and valleys"?
1. Company Overview: From Consumer Storage Giant to AI Infrastructure Rising Star
Sandisk Corporation (NASDAQ: SNDK) is a global leader in flash storage solutions, founded in 1988 by Eli Harari and others, headquartered in California, USA. The company was acquired by Western Digital in 2016, and was spun off in February 2025, relisting as an independent company.
· Core business: Based on NAND flash technology, it provides storage solutions covering three major areas: data centers, edge computing, and consumer markets, including solid-state drives (SSDs), memory cards, USB flash drives, and embedded storage products.
· Market Position: As one of the world's top five NAND flash memory suppliers, SanDisk holds over 11,000 patents and has driven the development of industry standards such as SD cards. Notably, through its joint venture with Kioxia, SanDisk has secured nearly one-third of the world's flash supply at a relatively low cost.
2. Core Positive: The "Printing Machine" Model Amid the AI Wave
SanDisk's recent explosive growth in performance is entirely due to the exponential driving force of AI in storage demand.
1. Performance "explosive," gross margin surpasses NVIDIA
In the fourth quarter of fiscal year 2026, SanDisk delivered a record-breaking report: revenue of $8.965 billion, a year-on-year increase of 372%; GAAP net profit reached $6.903 billion, compared to a loss in the same period last year. Its Non-GAAP gross margin soared to an astonishing 84.6%, even surpassing AI chip giant Nvidia and setting a new record in the NAND industry. The market attributed this to a severe shortage of NAND flash memory chips.
2. Strategic Transformation: AI data centers become the core engine, long-term contracts lock in future revenue
SanDisk's business focus is rapidly shifting from the consumer market to the high-value enterprise market. Data center business revenue this quarter was $2.977 billion, a year-over-year surge of 1298%, with its shipment bits accounting for 38% of total shipments soaring from 12% a year ago to 38%. More importantly, SanDisk signed 10 long-term agreements with 8 customers through the "New Business Model" (NBM), locking in minimum contract revenue of up to $93.9 billion, with financial guarantees of $16.5 billion. This model aims to transform SanDisk from a highly cyclical chip wholesaler into an infrastructure provider with stable cash flow.
3. Potential Positives and Negatives: The Market Votes with Its Feet, Fearing the "End of the Cycle"
Despite the impressive financial report, SanDisk's stock price plunged after the report, nearly halving its historical high, which deeply reflects two core concerns in the market.
1. Price increases are driven by price increases rather than demand growth, signaling a cycle has peaked
The root of market concerns is that two-thirds of this quarter's performance growth came from product price increases rather than substantial shipment increases. Currently, the high prices of memory chips have begun to hurt downstream demand. PC manufacturers raised prices due to rising costs, causing shipment declines, and smartphone manufacturers strongly resisting price hikes. Analysts point out that when the industry's profit margins are above 70%, this is often a sign that industry prosperity is approaching its peak.
2. Lack of high-end barriers may make it difficult to escape the "boom-bust" cycle
Unlike Samsung and SK Hynix's absolute technical barriers in HBM (High Bandwidth Memory), SanDisk's global share of the enterprise SSD market is only 2-3%, making it a market follower. Institutions like JPMorgan believe SanDisk's current high profits reflect more of the industry's cyclical prosperity than of structural improvement. As major suppliers resume capacity expansion and 3D NAND technology upgrades, the industry is expected to return to an oversupply "boom-bust" model starting in 2027, at which point SanDisk's ultra-high gross margin may be difficult to maintain.
SanDisk is at a critical stage of transforming from a consumer storage brand into an AI storage infrastructure provider. In the short term, the supply-demand mismatch brought by AI has earned huge profits; In the long term, whether it can break the inherent cyclical curse of memory chips and build a true technological moat will determine whether this is the prologue to the "king's return" or the "final celebration"#AI基建融资升温, Nvidia and Intel are diverging in their paths
Recently, financing in the AI infrastructure sector has surged, with NVIDIA and Intel taking completely different fundraising paths, and their strategic differences directly affect capital market performance.
Nvidia, in partnership with top financial institutions such as BlackRock, BlackRock, and Goldman Sachs, has built a dedicated AI computing power financing platform, planning to leverage over $500 billion in external funds. This money is not for Nvidia's own use, but is lent to downstream companies to purchase Nvidia GPUs and build new data centers, leveraging external support to boost its own hardware sales. However, after the announcement, Nvidia's stock price dipped slightly, and the market worries that massive credit volumes could increase industry debt risks, with corporate debt repayment pressure potentially passing on to hardware demand.
In contrast, Intel chose to raise its own funds for expansion, aiming to raise the scale of its share issuance to $20 billion, with market subscription demand exceeding $100 billion. All the funds raised will be invested in self-developed AI chips and factory construction, using equity dilution to strengthen its capacity and technological barriers.
Both models have their pros and cons: Nvidia leverages massive orders with light assets and expands faster, but depends on industry credit cycles; Intel's heavy-asset self-development makes operations more stable, and short-term equity dilution will suppress valuations. Overall, the influx of massive capital signals that AI infrastructure is still at the peak of expansion, but industry leverage is rising, and future oversupply and debt risks have become key concerns for investors, leading to more cautious market scrutiny of tech stock valuations.$BTC Is the sentiment really this bad now? Yesterday I saw BTC stagnating, and I thought maybe all the funds were waiting on the CPI. But the CPI hasn't even been released yet, and it already dropped below 64,000 last night? I guess some institutions are betting on the CPI exceeding expectations.
1. This week's market mainly depends on tomorrow's CPI data: the expectation is a year-over-year 3.42%, core 2.52%. If it's below expectations, it could mean no rate hike or even a rate cut, and BTC would go up; if it exceeds expectations, it will break below 62,000.
2. Institutions had net inflows yesterday, indicating some funds are withdrawing, probably because some institutions speculate the CPI data will exceed expectations. The current fear and greed index is 31, sentiment is positive, so retail investors probably won't run.
3. Regarding the CPI data and last week's non-farm payroll data, some friends asked if there could be falsification. It's actually quite possible, but consider this: government falsification is also to serve monetary policy. Non-farm payroll and CPI data themselves are meant to serve normal monetary functions, so even if falsified, interpreting policy from the data is still reasonable.
But I don't recommend betting on the CPI in advance, because retail investors' information sources are still much worse than institutions. Don't turn investing into a game of betting on size or luck. #本周三CPI公布，9月加息定价会改写吗？ #AIInfraEarningsWatch #CPIToResetFedBets #AIInfraFundingDiverges The $16.8 billion capital expenditure for the Terafab project in Texas has raised concerns about high cash consumption and rising macro capital costs. While self-developed manufacturing is extending the investment return cycle, it has also repriced market risk appetite.
The initial investment of $16.8 billion is concentrated in Terafab's self-developed AI chip factory in Texas, pushing the competition for computing power from direct leasing to a heavy capital investment stage on the manufacturing side.
The core drivers of market pricing are, in order: cash flow squeeze from heavy asset investments, expectations of self-developed chips reducing long-term computing power costs, and liquidity risk premiums during the construction cycle.
The upside scenario is based on the assumption that capital expenditures are efficiently converted into capacity. If trial production progress meets expectations and sustained demand can absorb this $16.8 billion fixed asset investment, the market's discount on heavy asset consumption will shift to a repricing of vertical integration premiums.
The variables to watch under this scenario are the efficiency of chip capacity ramp-up and the reduction in unit computing power cost, with failure signals being capital expenditures continuously exceeding budget and failure to achieve mass production within the scheduled timeframe.
The downside scenario is triggered by a high inflation environment and long-term cash consumption. When $16.8 billion in capital occupation causes liquidity tightness, or when R&D cycles are extended and returns on capital fall short of expectations, short-term positions will quickly reduce the risk appetite premium in the technology sector.
The variable to watch under this scenario is the speed of free cash flow deterioration and the suppression of macro interest rates on heavy asset expansion. The failure signal is that external financing channels are smooth and high-margin businesses should promptly fill cash flow gaps.
If macro risk appetite contracts sharply or alternative technology paths emerge in the computing power manufacturing chain, the current assessment that $16.8 billion in heavy asset investment could be exchanged for long-term competitive barriers will completely fail.
In the next seven days, it is important to closely monitor fluctuations in macro funding rates after the massive capital expenditure is confirmed, as well as the rebalancing trends of institutional positions in heavy technology asset assets.
#闪迪8月13日投资者日临近, #霍尔木兹海峡通航协议未落地 of divergences in financial reports, oil price risks have risen #火箭实验室财报超预期, and the commercial space boom continuesShort-term BTC takeaway 📉📈
This post says Hormuz is currently a macro risk for BTC, but it can work both ways.
Bearish scenario: Hormuz tensions → oil ↑ → inflation expectations ↑ → Fed cuts become harder → DXY/yields ↑ → liquidity ↓ → BTC pressure.
Bullish scenario: Hormuz deal/reopening → oil risk premium ↓ → inflation pressure ↓ → easier Fed expectations → liquidity improves → BTC could benefit.
What to watch
🛢️ Brent oil
🌍 Hormuz negotiations
💵 DXY
📊 U.S. Treasury yields
₿ BTC support/resistance + volume
My take: Don't short BTC solely because of Hormuz. The stronger bearish confirmation would be oil rising alongside DXY and yields while BTC loses key support. If Hormuz tensions ease and oil falls, the same macro setup could quickly turn bullish.Hormuz Deal Unresolved: Oil and Crypto Stand at a Critical Crossroads
Hormuz remains unresolved. While negotiations between the U.S., Iran, and Oman have made progress, disagreements over shipping routes, transit fees, and passage conditions mean geopolitical risks have not disappeared.
Brent has climbed to around $84.95 per barrel, showing that markets are still pricing in a geopolitical risk premium linked to Hormuz.
This matters significantly for Crypto:
Hormuz tensions → Oil rises → Inflation expectations increase → Fed easing becomes harder → USD/yields rise → Risk-asset liquidity weakens → $BTC and Crypto face pressure.
Conversely, if Hormuz reopens sustainably, the geopolitical premium could decline, oil could cool, and monetary-policy expectations could improve — creating more room for $BTC and the broader Crypto market to recover.
Investors should therefore watch Hormuz, Brent, the U.S. dollar, Treasury yields, and $BTC price structure together.
The key takeaway: Hormuz remains unresolved, so the risk has not disappeared. A durable agreement could become a positive catalyst for risk assets, while a breakdown in negotiations could quickly trigger another wave of volatility.
If you find this information useful, follow me to stay updated and discuss the latest developments across the Crypto market and Wall Street.
#HormuzDealUnresolved
#StrategySellsBTCAgain
#BTCETHETFFlowsDiverge
$BTC
$ETH H #AIInfraEarningsWatch #CPIToResetFedBets #AIInfraFundingDiverges "Say Things Without Saying Anything, Say Things Hard When Nothing Matters"
🦉 When I saw news of large ETF inflows, my first impression was definitely not that the market was about to take off, but that institutional funds returning only represented liquidity recovery, not that incremental funds would immediately enter and rally the market.
Last week, US spot Bitcoin ETFs saw a total net inflow of $865 million, ranking among the top inflows in nearly 15 weeks. BlackRock IBIT alone attracted nearly 700 million in funds, supporting the vast majority of inflows; ETH also performed impressively, with a net inflow of $244 million, marking five consecutive weeks of net inflows. This clearly shows that the window for institutional capital entry has loosened once again.
🐢 However, a rational cooling and review of the market is necessary; the deep correction in July remains fresh in memory. Currently, the coin price continues to fluctuate in the $64,000–$65,000 range, and overall market sentiment remains wavering and wait-and-see, with a mix of suspicion and suspicion. The return of ETF funds only proves that mainstream institutional funds dare to test positions with small positions; spot market trading volume and retail investor sentiment have not warmed up in tandem.
Macro liquidity is the core driver determining the upper bound of the market. Only when the Fed maintains a wait-and-see approach to monetary policy and weakening nonfarm payroll data can lower rate hike expectations, risk assets can have room to recover and breathe; Once interest rate expectations shift, capital in the market can quickly flee.
Ethereum has been increased by institutional funds for five consecutive weeks, signaling that institutions are reassessing Ethereum's long-term value and treating it as the second core allocation target in the crypto sector. However, the rotation of Bitcoin leading the rally and Ethereum following the rally requires volume support; relying solely on daily ETF fund broadcasts to drive sentiment is far from enough.
🦅 My personal view is straightforward: capital inflows are a positive signal, but definitely not a signal for a large-scale market entry.
First, observe whether ETFs can sustain continuous inflows, and second, confirm that the spot market is genuinely strengthening support. Before these two conditions are met, strictly control positions and avoid being swayed by trending market trends.
Capital flows reveal market direction, but cannot bear the risk of losses for traders. $BTC $ETH #AI基建融资升温, Nvidia and Intel's paths diverged. #本周三CPI公布, will the September rate hike pricing be rewritten? #现货ETF资金分化, BTC selling pressure remains Strategy Sells BTC Again: What Signal Is the Whale Sending?
Strategy has sold Bitcoin again — and the market should look beyond the 1,690 BTC figure.
During August 3–9, Strategy sold 1,690 $BTC worth approximately $108.6 million, at an average price of $64,262 per BTC. The proceeds were used to repurchase approximately 1.15 million STRC preferred shares.
One week earlier, Strategy sold another 1,638 $BTC , generating approximately $104.7 million. In two weeks, the company sold more than 3,300 BTC, worth over $213 million.
Yet Strategy still holds approximately 840,447 $BTC , with a total cost basis of around $63.36 billion, or roughly $75,385 per BTC. Its U.S. dollar reserve has increased to approximately $4.65 billion.
This looks more like a liquidity strategy than abandoning Bitcoin.
Strategy is converting BTC into liquidity to strengthen its balance sheet and repurchase STRC, while also raising approximately $653.1 million through MSTR share sales.
But one signal cannot be ignored:
Strategy has gone several weeks without buying Bitcoin while continuing to sell BTC.
If this continues, the market will ask:
Is institutional demand taking a temporary pause, or are corporate Bitcoin strategies entering a new phase?
Selling 1,690 BTC remains small compared with its 840,447 BTC holdings. It does not prove Strategy has turned bearish.
The real signal is frequency.
Markets trade not only on supply, but on the belief that large buyers will absorb it.
When a major corporate Bitcoin holder shifts from “buy BTC” to “optimize liquidity,” sentiment adjusts.
Strategy has not turned its back on Bitcoin. But during volatility, liquidity can matter as much as conviction.
If $BTC remains under pressure while institutional demand fails to return, the risk could extend beyond a single sale.
But if Strategy stops selling and resumes accumulation, it could signal renewed institutional confidence in Bitcoin.
Watch what Strategy does next — not just what it did today.
$BTC #AIInfraEarningsWatch #CPIToResetFedBets #AIInfraFundingDiverges The 500 billion yuan "computing power credit platform" jointly developed by NVIDIA and Wall Street has completely upgraded the AI market's gameplay.
Simply put: Wall Street pays the price, companies use GPU mortgages to buy computing power, and Nvidia provides a 25% back.
The signal is very clear:
1️⃣ Assetization of chip finance: GPUs have officially transformed from "electronic consumables" into interest-bearing assets that generate cash flow.
2️⃣ Unlocking cash flow from major clients: giants don't need to crowd their balance sheets, CapEx becomes long-term leases, and buying can't stop.
3️⃣ Shedding the burden of "circular financing": Let independent financial institutions underwrite to isolate compliance risks.
Hardware sales ➡️, financial leverage ➡️, and operating commissions—Jensen Huang sold computing power into infrastructure leasing business.
But it's important to note that once leverage is maxed out, the entire industry is betting on whether downstream AI monetization speed can outpace debt interest. Are you optimistic about this model?
#NVIDIA #AI #英伟达 #美股The market isn’t moving on one headline today. It’s a tug-of-war between institutional positioning, Ethereum’s next phase, tightening liquidity and a macro backdrop that could stay restrictive. Here’s what matters 👇 🟠 $BTC — Institutional selling meets structural demand $BTC is around $63.9K, down ~1.5%, while $ETH sits near $1.87K, down ~2.3%. Strategy sold 1,690 BTC for ~$108.6M between Aug. 3–9, taking its holdings to roughly 840,447 BTC. The proceeds are earmarked for its STRC repurchase pAI is entering a new phase: the market's focus has shifted from "whose model is stronger" to "who can turn computing power into sustained cash flow." 📈 The latest market news shows that NVIDIA is collaborating with several major financial institutions to promote AI computing infrastructure financing platforms, aiming to leverage over $500 billion in third-party capital. This means that AI computing power competition is no longer just a capital expenditure game for tech companies, but is gradually turning into a super infrastructure investment cycle involving technology + finance + energy + data centers. (GuruFocus) What's even more noteworthy is that Wall Street has begun to re-examine AI's massive capital expenditures: the real question is no longer "whether AI should keep burning money," but whether these investments can ultimately be converted into revenue, profit, and stable cash flow. This change is also happening in the crypto market. The era of sweeping rallies, once driven by concepts, sentiment, and liquidity, is fading, and the market is entering a true value screening cycle. The future will no longer be about whose story is the sexiest, but who can sustainably attract capital, have real users, and form a commercial closed loop, and who will achieve higher valuation premiums. 🔥 AI is reshaping global capital flows: In the past, capital was mainly allocated around GPUs, servers, cloud computing, and large models. Now, capital is starting to seek opportunities along the entire industry chain: ✅ AI chips and HBM
✅ Data centers and cloud computing
✅ Power and energy infrastructure
✅ Networking and optical modules
✅ AI models and AgenTEXAS IS AUDITING DATA CENTERS — COULD THIS BECOME A HIDDEN CATALYST FOR $BTC ?
Hundreds of large projects are competing for access to the Texas power grid.
AI, data centers, and Bitcoin mining all depend on electricity — but grid capacity is limited.
Texas is now auditing data-center applications for ERCOT connections. The scale is massive: about 474 GW of large-load requests, with roughly 90% tied to data centers — over 5x Texas’ peak demand.
Meanwhile, Bitcoin trades in a tight range:
$BTC: ~63,940 USDT24H High: 65,368 | Low: 63,818Volume: ~3.28K BTC (~211M USDT)
On the short-term chart, BTC sits near 63,940, below key moving averages around 63,970–64,051, showing weak momentum.
But the bigger story is off-chart.
If new data centers face delays, existing sites with secured land, substations, and power access become far more valuable.
Some Bitcoin miners already control this infrastructure.
Bernstein notes that limited new power capacity could boost the value of existing mining and AI-ready sites, giving them a structural advantage.
This shifts the narrative:
Bitcoin mining sites may evolve into AI infrastructure hubs.
And the key asset becomes not hardware — but electricity access.
This doesn’t directly push $BTC higher from ~$63,940. The audit doesn’t change Bitcoin demand or ETF flows.
Near-term impact is more likely on mining equities and infrastructure valuations.
But the longer-term trend matters.
As AI demand surges, electricity becomes scarce, and some miners may redirect capacity from Bitcoin to AI/HPC workloads.
So the real chain may be:
AI boom → power scarcity → tighter grid access → rising value of existing energy infrastructure → miners become strategic power holders.
BTC remains around $64,000, but the deeper competition is already forming — not for coins, but for megawatts.
In the AI era, the most valuable resource may not be compute.
It may be electricity.
$BTC
#DailyOrbit
#OKXOrbitTopics #AIInfraEarningsWatch #CPIToResetFedBets #AIInfraFundingDiverges I don't think I am alone when I say that many people are waiting to load up with spot positions at the moment. Historically, $BTC should bottom around October this year, and this time around it feels like everyone is trying to time the market.
Recently we've had a lot of news about Saylor selling, but even that won't move the price downwards anymore. Also, the Clarity Act doesn't seem very likely to be approved this year, so we still have to wait for that one. The Clarity Act should be bullish for big assets like BTC and ETH, and should help with momentum (all boats will be lifted by the tide).
August and September are historically the worst months for $BTC, and if we get a bigger flush, then this dip is probably worth buying for good alts (I have my list in the pinned, but HYPE, LIT, ZEC, PUMP++ are some of them).
Almost all altcoins trend down long-term, your job during the next months is to try to allocate into the ones that hopefully don't.
$BTC
#AIInfraEarningsWatch
#AIInfraFundingDiverges 🚨 CRYPTO LIQUIDITY CHECK — Aug 11, 2026 🚨
Volumes are drying up FAST. Daily spot trading is down ~70% from January's highs, sitting near $15-20B/day. Just 6 exchanges now control 60%+ of all spot activity. Liquidity isn't spreading — it's consolidating.
📉 THE DAMAGE (24H)
$BTC → $63,911 (-1.44%)
$ETH → $1,871 (-1.96%)
$XRP → $1.0115 (-1.69%)
$SOL → $75.95 (-0.35%) 🛡️ holding the line
Why? Not crypto-native. Iran just signaled the Strait of Hormuz stays shut longer than hoped. Oil spiked, risk-off hit everything from equities to Bitcoin in lockstep.
💵 STABLECOINS = THE REAL TELL
Total stablecoin cap: ~$287-290B (down ~5%)
$USDT still king: ~58% share, ~$185B
$USDC: ~$78B
TRON's USDT supply just hit an ALL-TIME HIGH of $87.9B, moving $2.1 TRILLION last quarter alone. It's not hype — it's the dollar rail for entire economies now.
🔮 WHAT TO WATCH
→ Oil above $85/barrel for a week = prolonged risk-off
→ CLARITY Act vote (delayed past Senate recess)
→ Brazil's Oct licensing deadline ($319B market)
This isn't a crypto breakdown — it's a liquidity squeeze wearing a geopolitical mask. 🌊
#Bitcoin #Crypto #Liquidity #Altcoins
#AIInfraEarningsWatch #CPIToResetFedBets Spot gold surged strongly above $4,400, reaching an intraday high of $4,435, marking a two-month high. Supported by geopolitical uncertainty, safe-haven funds continued to flow into hard currency assets like gold.
Currently, the U.S. CPI has not yet been released, and the market has already priced in the Fed's policy expectations in advance. Previously, weaker nonfarm payroll data led the market to lower bets on rate hikes, causing both U.S. Treasury yields and the dollar to weaken simultaneously, which directly benefited precious metals.
CPI is the core watershed for the upcoming market: if inflation data cools and rate cut expectations further ferment, the dollar and interest rates come under pressure, gold continues to rise, and liquidity-sensitive crypto assets like BTC and ETH will see capital flow back; Conversely, if CPI strengthens beyond expectations and hawkish sentiment rebounds, short-term fluctuations in major assets will sharply amplify.
This round of macro market activity has just begun, and gold is the first to show a trend. Once the CPI releases a dovish signal, funds will widely spread to various risk assets.
Risk warning: Only market strategies are shared, do not constitute trading advice, no inappropriate guidance, comply with community conventions. $BTC $ETH $XAU #本周三CPI公布, will the pricing for September rate hikes be rewritten? The 30-year U.S. Treasury yield has risen to a staggering 5.25%, basically reaching the level on the eve of the 2007 financial crisis!
Moreover, U.S. national debt has officially surpassed $40 trillion, marking a historic first! It took only 150 days to go from $39 trillion to $40 trillion.
At the current interest rate of 3.75%, the annual interest expenditure on U.S. debt reaches $1.5 trillion. This means that for every $5 in taxes collected, $1 is directly used to repay interest.
The current situation is absurd:
Raising interest rates, cutting rates, shrinking the balance sheet, expanding the balance sheet are not the same—each path has fatal side effects.
If the U.S. cannot quickly resolve the U.S.-Iran conflict and calm oil prices and inflation, it will ultimately have to rely on classic liquidity injections, because soaring long-term bond yields will seriously threaten U.S. stock valuations.
The only way to break this must be to aggressively tax the entire world!On August 11, the total crypto market cap was about $2.2 trillion, BTC dominance reached around 58.9%, and ETH's share dropped to just 10.2%—a figure cut in half from ETH's 18% peak during DeFi Summer 2021. Many people's first reaction was that ETH was failing, but the truth is more complex: it's not ETH falling, but the market is repricing.
BTC dominance has climbed from 52% in Q1 2025 to about 59% now, with the core driving force not BTC's surge, but funds seeking safe havens. Global geopolitical conflicts, tariff games, and Fed policy swings have driven institutional funds into BTC ETFs seeking a safe haven for "digital gold." Spot BTC ETFs have attracted over $56 billion cumulatively, with BlackRock's IBIT alone accounting for $54 billion+. This structural buying has directly propped BTC dominance above 50%—the longest since 2017. More importantly, if you exclude the more than $300 billion in stablecoins from total market value, BTC's control over real venture capital is actually close to 64%, making the market far more "BTC-like" than it appears.
ETH's share shrinks to 10% not because it's falling—it's falling too, but even harder than BTC. Since its October 2025 high, BTC has fallen from $126,200 to $59,000, a 53% decline; ETH has fallen 67% over the same period. The ETH/BTC exchange rate crashed from 0.088 in December 2021 to 0.030, depreciating 65% relative to BTC over four years.
Four structural issues are squeezing ETH at the same time. L2 backlash on mainnet: Arbitrum, Optimism, and Base have taken most of the trading volume, but the fees for mainnet return are negligible. Base made 94 million in profit but only paid 4.9 million blob fees to mainnet. Mainnet fees, which once earned $30 million daily, are now down to about $500,000, with ETH turning from deflation back to mild inflation. Solana Grabs Territory: In Q1 2026, Ethereum's on-chain revenue dropped to fourth, behind Solana, Tron, and BNB Chain. Solana confirms in 400ms vs. ETH blocks in 12 seconds; meme coins and high-frequency DeFi all run to Solana. Institutional funds are 7:1 biased toward BTC: BTC ETF size is about $128 billion, ETH ETF only about $18 billion. Large funds buy crypto first, buy BTC; ETH is the "if you have a budget" option. Glamsterdam upgrade delayed repeatedly: the parallel transaction processing and 78% gas fee cuts originally planned for H1 2026 may now be pushed to Q3 or even Q4. Each delay is giving competitors time.
But ETH has a history of comebacks. In September 2019, ETH dominance also fell to around 10%, but over the next 18 months, ETH quadrupled against BTC, doubling its share to 20%. The catalyst at that time was the DeFi boom. Now there are three potential catalysts: the successful launch of Glamsterdam, the Federal Reserve's rate cut in the second half of the year, and the approval of ETH ETFs with staking yields. ETH/BTC is historically a key support in the 0.028-0.030 range; if it holds and rebounds, ETH's share could return to 12-13%; If it falls below 0.028, the next support is at 0.020, corresponding to ETH dominance dropping to 7-8%, and Solana really has a chance to challenge the second-place position.
The current market is not a simple binary opposition of "$BTC good, $ETH bad." It is that under macroeconomic uncertainty, funds have chosen the hardest and least explanation-needing assets. BTC's narrative is "digital gold"—simple, easy to understand, and recognized by institutions. ETH's narrative is "the world's computer," but L2 fragmentation, the rise of competitive chains, and upgrade delays increasingly require "explanation costs."
BTC dominance approaching 60% is not the end, but a signal: among risk assets, the market is making the most conservative choices. ETH's turnaround is not about BTC falling, but about a new narrative breakthrough—like the DeFi Summer in 2020—that makes funds willing to take on the risks of "smart contract platforms" again. Otherwise, "BTC-ization" will continue until a macro liquidity inflection point or a true killer application emerges in the ETH ecosystem.