Orbit Post Sitemap
Some thoughts on NVIDIA and Wall Street's $500 billion financing guarantee:
- This should help prevent GPU sales from being impacted by cash flow tightening among hyperscale cloud providers, who may no longer be able to fully prepay for GPUs.
- It should accelerate the expansion of the GPU market TAM; I don't see this as circular financing.
- Calling this circular financing is like criticizing the introduction of auto financing, saying car companies are essentially buying their own cars.
- Another point worth noting is that Wall Street is increasingly starting to view GPUs as recyclable collateral.#霍尔木兹协议待落地，原油风险等待定价 My response is "Don't get excited yet"
As soon as this news came out, the group chat got lively again. The negotiations at the Strait of Hormuz failed, with the US and Iran pressuring each other, and the situation suddenly became tense again. This kind of geopolitical risk always triggers the market's first and most direct reaction: oil prices will rise, safe-haven assets will rise, and risk assets will shake.
But after looking at the market, I'm not that worried.
The Strait of Hormuz is the main artery for global oil transportation. When the Middle East situation tightens, oil prices definitely have upward momentum. But the key is to see: will the conflict actually affect supply? Currently, it looks more like an "escalation of pressure" stage, not a real supply cut. So oil prices will rise, but whether it can sustain is questionable.
What impact does this have on us?
First, rising oil prices will push up inflation expectations. This connects well with the earlier point about "inflation expectations not falling but rising." The US is already worried about a price rebound, and if oil prices push higher, the Fed's room to cut rates will be further squeezed. This is not good for risk assets.
Second, safe-haven sentiment heats up, and gold will benefit in the short term. $BTC sometimes follows safe-haven trends, sometimes risk trends, but recently it has been more tied to macro liquidity. If oil prices push inflation higher, the Fed won't dare to ease, and Bitcoin will be suppressed. So the safe-haven logic doesn't necessarily apply to $BTC.
Third, US stocks and crypto assets may come under pressure. Geopolitical risk + rising oil prices = stagflation concerns. Under this combination, growth stocks and risk asset valuations will be suppressed. If oil prices continue to surge, $BTC and $ETH will likely fall back.
Here’s my personal view.
I still hold long positions in $BTC, but the position size is not heavy. After seeing the news last night, my first reaction was not to add positions but to check stop-loss levels. Geopolitical events come fast and may go fast, but the volatility in between can be very intense. I don't want to bet on whether it will be bullish or bearish because the logic can change anytime.
I pay more attention to changes in inflation expectations. If oil prices keep rising because of this, the previous expectation of "no rate hike in September" will be diluted, and the market may reprice "higher rates for longer." This is unfavorable for $BTC's medium-term trend. Conversely, if it's just a short-term emotional shock and oil prices rise then fall back, the market impact is limited.
My approach is:
Don't chase gold, and don't add to $BTC positions. First observe if oil prices can hold steady, then watch for follow-up statements from Fed officials. I continue to hold my $BTC position but have set clear defense levels. If $BTC falls below $62500, I will reduce some to prevent a chain reaction of selling caused by escalating geopolitical risk.
At times like this, the worst thing is to chase news back and forth. Geopolitical news changes too fast; you chase in today, and it might reverse tomorrow. I choose to reduce trading frequency and keep my position size at a level that lets me sleep well.
In summary: failed negotiations are not the end of the world, but don't treat it as an opportunity to rush in. Oil price is the core variable, inflation is the real anchor. Wait for the situation to become clearer before deciding to add or reduce.
$BTC $ETH
#霍尔木兹协议未落地，油价风险再升温？
#标普盈利超预期，华尔街为何仅看7894点 Very soon! The U.S. national debt is approaching $40 trillion.
And 42% of the income tax collected by the U.S. government is used to pay interest on the national debt.
The U.S. government's debt grows by $2.1 trillion every year. Most and even more of the U.S. government's budget is swallowed up by interest payments.
If the Federal Reserve raises interest rates, the situation will worsen further.
There are only two ways in the world to reduce debt: 1, war; 2, let inflation run wild for a while, or even longer...
The U.S. just issued 30-year Treasury bonds with a yield of 5.22%, the highest borrowing cost since 2001. And two days ago, the 10-year Treasury yield hit a new high not seen since 2007.
With this situation, do they still dare to raise interest rates?
Moreover, can gold be pushed even lower?Torn Reality: Consumption Has Slumped, but Inflation Expectations Have Risen
July retail data and August consumer confidence both fell, confirming the stalling of the consumption engine. Cooling inflation, loosening employment, and weakening consumption—these three major signals combined sharply reduce the necessity for a rate hike in September (CME's probability of no rate hike has risen above 67.5%).
But this is exactly where the market is most conflicted: one-year inflation expectations have instead risen counter to the trend to 4.3%. While people are tightening their wallets, they still expect prices to rise. Under the shadow of this "stagflation" expectation, the Federal Reserve finds it difficult to aggressively cut rates, long-term interest rates remain capped, and risk assets should not expect a quick easing.
⚡ Reflected in the current state of $BTC
Weak consumption has lowered rate hike expectations, giving the market a short-term breather. But before inflation expectations die down and liquidity is truly released, the $65,000 level will likely continue to test patience repeatedly. Don't expect a turnaround from just one or two data points; before the September policy meeting, this is the most mentally draining "garbage time."
#消费动能转弱，9月政策仍受通胀制约 #比特币BIP-110 fork stalled, miner support insufficient #Spot ETF funds diverge, BTC selling pressure remains #Risk heating up, why BTC can't gain safe-haven benefits⚠️
Tensions in the Middle East escalate again, Trump makes statements on Iran, US-Iran confrontation intensifies, Hormuz navigation talks break down, global risk-off sentiment rapidly rises.
But the market reveals a harsh truth:
When a real crisis hits, BTC does not act as a safe-haven asset.
Market capital flows are straightforward: when risk erupts, funds rush first to traditional safe havens like gold and US Treasuries, while highly volatile crypto assets are sold off first.
"Digital gold" is more a bull market narrative.
When panic truly sets in, BTC is often the first risk asset to be reduced.
This is the fundamental reason why BTC failed to rally as a safe haven during this round of geopolitical escalation. Short term caution is needed; if the conflict continues to escalate, the crypto market is likely to face pressure.
However, it's not all bearish:
If the conflict keeps pushing oil prices higher, further driving inflation, it will directly constrain the Fed's pace of rate cuts.
The likely market path is: BTC first suffers a sentiment-driven sell-off, then the market re-prices the long-term logic of US dollar credit and global liquidity reshaping, which will bring opportunities.
Core summary:
👉 No short-term safe-haven gains, long-term play on liquidity dividends
Do not blindly enter to speculate on safe-haven moves driven by geopolitical risks.
#Hormuz navigation talks fail, US-Iran pressure escalates
$BTC $ETH $SNDK🚨 A staked $ETH ETF just revealed a liquidity problem traders need to watch.
21Shares reported $48.4M in redemptions during H1 2026, while 86.42% of its $ETH was staked at quarter-end.
The issue is timing.
Staked $ETH cannot be moved instantly, creating a potential mismatch between ETF redemptions and available liquid $ETH.
There was no failed redemption.
But if outflows accelerate, this could become much more important for U.S. $ETH ETF markets. 👀On the market, the Pre-IPO contract trends of $ANTHROPIC and $OPENAI almost overlap, but their fundamental ledgers are diverging.
Anthropic's Q2 revenue reached 11.5 billion and doubled quarter-on-quarter, while also turning profitable; OpenAI's annualized revenue hits 40 billion, but its profit expectations have been postponed to 2030.
The nearly twofold valuation gap between the two reflects the secondary derivatives market's emphasis on cash flow realization ability over scale growth.
The high premium is currently based on extremely optimistic pricing expectations, and the fundamental divergence is gradually shrinking investors' willingness to pay for pure narratives.
If subsequent disclosed data further confirms improved profit quality, contract pricing is expected to reopen valuation space after absorbing selling pressure, while shrinking trading volume will signal a weakening of this logic.
If future performance cannot maintain above-expectation growth, combined with the thin liquidity characteristic of the Pre-IPO market, prices are prone to irrational downward adjustments under selling pressure.
When market sentiment shifts back to computing power expansion rather than immediate profits, the current pricing logic based on profitability will be overturned.
The most important variable to watch in the coming week is whether the price gap and premium rate between the two targets converge in the absence of new earnings catalysts.
#闪迪投资者日后股价大涨，长期目标待验证 #海力士扩产提速，资本开支能否兑现回报 #霍尔木兹协议待落地，原油风险等待定价
Over the weekend, these two lines moved in sync but in completely opposite directions. On one side, Iranian Foreign Minister Araghchi held a press conference in Tehran, saying the Iran-Oman Hormuz transit plan is "very close." On the other side, Trump shouted at a rally in New York, "After defeating Iran, I will soon declare the Strait of Hormuz as U.S. territory." One side is working on a transit plan, the other is shouting sovereignty—the crude oil price climbed back to 88/84 this week. It's unclear whether this reflects expectations of the "agreement landing" or the "agreement not landing."
The Oman agreement itself does not resolve the crisis. Araghchi himself revealed that reaching an agreement does not mean the strait will reopen; Iran no longer accepts the old Traffic Separation Scheme (TSS) and wants a new plan. More glaringly, IRGC spokesperson Mohebbi said the reopening of the strait "has nothing to do with Iran-Oman negotiations" and is Iran's own mechanism. This means the Foreign Ministry is negotiating, but the Revolutionary Guards do not acknowledge it. Iran's Supreme National Security Council Chairman Zolghadr, also an IRGC commander, directly links "opening" to "U.S. correcting its behavior"—the usual stance is no negotiation. @qinbafrank and @PhyrexNi on X saw through this early: the "Hormuz transit plan" is not about opening the strait but mainly about "how to charge fees compliantly."
Washington is also conflicted. In June, the U.S. and Iran signed a Memorandum of Understanding (MoU), which Iran says the U.S. violated; it expires Monday. The Omani Foreign Minister is mediating, but Trump was still shouting "declare sovereignty" over the weekend. CENTCOM Commander Brad Cooper toured six Gulf countries and boarded the USS Lincoln—blockade intentions remain. On Reddit, a report titled "Trump officials pressure Oman to accept Iran's terms" circulated, meaning the U.S. is pressuring Oman to accept Iran's conditions; Hacker News headline "gulf states accept a new normal: Iran is in control"—the Persian Gulf Arab states have accepted "Iran controlling Hormuz as the new normal," preferring to negotiate terms rather than fight again.
The market has already positioned pricing for the "agreement pending." Brent returned to $88, WTI to $84 this week. Twitter users like MarioNawfal track blockade news to catalyze short-term moves. But J.P. Morgan's global commodity strategist Natasha Kaneva projects a contrary mid-term path—Q3 Brent at 86, Q4 at 80, year-end back to 78; longer term, oil prices return to the 60-65 range in the second half of 2027. The reason is demand destruction is worse than expected, commercial inventories are insufficient, and China's permanent gasoline demand loss is about 180,000 barrels/day—the price hasn't surged because of Iran but because consumption has collapsed first. EIA's 17.4 million barrel inventory build this week is the other side of the same coin.
Another structural change is quietly being priced in. UAE exited OPEC in May, aiming to expand production to 5 million barrels/day by 2027; OPEC's control ability declines, losing 11% share by 2025. This means after "Hormuz reopens," supply recovery will be faster than anyone expects—JPM sees a 1.2 million barrel/day surplus as early as August, Q4 reaching 97% of pre-war levels. Short-term trades the blockade risk premium; mid-term trades the scissors gap between supply recovery and demand destruction.
The U.S. is using SPR to buy time—@PhyrexNi cites data showing strategic reserves have fallen below 300 million barrels, consuming 110 million barrels since the war began; @nytchinese summarizes the side effects—Pakistan's electric motorcycle sales surged, and China's electric vehicle exports soared 120% year-on-year in the first half. With high oil prices lasting, buyers on the demand side neither tolerate nor resist but adapt.
So the real hook of "agreement pending, risk waiting to be priced" is not whether the agreement is signed but two timing differences: whether the MoU expiring Monday will automatically extend, and when the Oman plan details will be disclosed. If smooth, the blockade risk premium will quickly retreat; if it repeats Iran's previous stance of "no ceasefire, no extension," all hedges near $88 Brent will flood back in. Are you betting the Oman plan details will surface this week first, or the MoU expiration news on Monday?
#CrudeOil #Hormuz #GeopoliticalRiskThe most common misunderstanding on the one-hour trending chart is that the total volume is mistaken for trends. The official snapshot of OKX Onchain OS from 06:00 on August 16 shows that BTC, ETH, and SOL were mentioned 43, 8, and 23 times respectively in the past hour; The total 24-hour volume was 1,374, 465, and 465 times. To compare the two windows, you can first divide the total of 24 hours by 24, then use the latest hour to compare. The results were BTC at 0.75x, ETH at 0.41x, and SOL at 1.19x. A score above one indicates activity in the most recent hour compared to the full-day average; below one indicates relative quiet; This is just a discussion of speed, not rate of return. By this logic, BTC has slowed down, ETH has clearly slowed down, and SOL has slightly accelerated. Whoever has the highest original mentions may not necessarily be the one whose baseline temperature is rising the fastest. Distinguishing between "the highest volume" and the "fastest acceleration" can reduce many misjudgments. The tone is another layer to consider. BTC is a mix of bulls and bears, with bullish and bearish rates of 40% and 35%, respectively; ETH is clearly bullish, with proportions of 63% and 0%; SOL is clearly bullish, with proportions of 61% and 4%. The key here is the denominator. ETH only happens 8 times per hour, SOL 23 times, so just a few new texts can significantly change the percentage; Although BTC has a larger sample, it may also include forwards and references from the same event. Ranked by percentageInfluencer tea drink stocks are not really buyable; investors are investing in the supply chain, but the valuation is driven by the "influencer" attribute, so shorting at the top has a very high success rate.
On the other hand, tea brands like Gu Ming, which many northerners haven't really heard of, are pure supply chain-strong companies whose stock prices have risen quickly.
Mixue Bingcheng shows that a good company is not necessarily a good stock; the valuation is too high.Consumption is also starting to falter.
Retail sales in July fell by 0.6% month-on-month, while the market had originally expected a 0.1% increase. Car sales are sluggish, online shopping has declined, and even gas station revenues are dropping along with fuel prices. The consumer confidence index fell from 55.2 to 51 in August, the first decline in three months.
With inflation cooling, employment loosening, and consumption weakening—several signals combined—the necessity for a rate hike in September is indeed diminishing. CME data shows the probability of no rate hike has risen to 67.5%, with some institutions even seeing 71%.
However, paradoxically, the one-year inflation expectation has risen from 4.2% to 4.3%. While people are tightening their wallets, they still expect prices to continue rising. Under this mindset, rate cuts won’t come quickly, and don’t expect risk assets to loosen up all at once.
Back to $BTC itself, weak consumption data has lowered rate hike expectations, offering a short-term breather. But inflation expectations haven’t come down, long-term interest rates remain high, and the 65000 level will likely continue to consolidate. Don’t expect a turnaround from just one or two data points; it’s a waiting game until the September meeting.
#消费动能转弱，9月政策仍受通胀制约 Guys, lately I've been watching Bitcoin's candlesticks, and I suddenly have the feeling — has this thing changed its temper? Back when the market crashed, it was so decisive and decisive—when a big bearish candlestick came down, people were so scared they could barely hold their tea steadily. But now, flapping back and forth in this range, the slashing force was noticeably gentler, like a raging cat suddenly learning to retract its paws. 😌 We can't just look at the lively surface; we have to take it apart and crush it to see the inside and its doors. Look at the participation rate of perpetual contracts—it's been declining all the way, while the spot market's trading volume has quietly raised the lower boundary. Translated into plain terms, this scene is—the main players shouting and dumping the market now are not actually holding the coins, but those who leverage and use "technology" in the futures market. Leverage is a market where the heart races; at the slightest sign of trouble, you immediately set a stop-loss order to break the market first. 💔 But the real highlight is that the group receiving the stock is incredibly calm. When the contract market drops down, they silently act like a sponge, gradually absorbing selling pressure. Whatever was smashed, it caught as much as it could, firmly holding the price of the roller coaster that was driving downward, preventing it from flying off the track or breaking through the floor. This scene was like a group of professional players playing chess against drunken men in a casino, with gamblers playing cards recklessly shirtless, while the pros calmly reinforced their defenses. 🧱 Honestly, the biggest fear in market sentiment isn't a drop, but chaotic panic. The current decline is orderly and supportive, which actually shows the underlying capital structure ratioAccount Position Divergence Radar
The side with more people does not necessarily have heavier positions; this chart specifically separates quantity and weight.
$DOGE overall accounts and top accounts are biased towards long positions, but the top holdings scale is biased towards short positions. The number of accounts and position weight are not aligned. The 15-minute increase in short positions during the decline indicates new positions participated in this downward pressure. The next step for the long side is not more accounts but confirmation of the top position weight.
$CAP account numbers consistently lean short, but the top holdings ratio is above 1, meaning the short-biased number of accounts has not translated into a top short position advantage. The rise was not accompanied by position withdrawals; new positions have added conditions for this trend to continue. If the price continues to weaken while the top holdings ratio remains above 1, this divergence has not truly converged.
$PEPE account numbers have already leaned towards the long side, but the top position scale has not followed. The current divergence comes from quantity and weight. The price-position combination falls into increased short positions during the decline, making the short-side pressure easier to continue, but it still depends on whether the price continues to break lows. The account side is already biased long; next, it depends on whether the top positions are willing to push the weight to the same side. 📊 $SNDK Contract Liquidation Update (August 15)
According to liquidation data, the whale successfully executed a textbook short squeeze harvest on SNDK from short to long cycles. Shorts controlled the market from the 1-hour mark throughout, with total liquidations exceeding $310,000.
Time Total Liquidations Long Liquidations Short Liquidations
1 hour $15,800 $677.89 $15,100
4 hours $30,200 $5,843.44 $24,300
12 hours $143,500 $11,100 $132,400
24 hours $315,000 $29,500 $285,500
From the $SNDK liquidation data, 1-hour short liquidations crushed longs, with shorts 22.3 times the longs, unleashing a nuclear-level short squeeze intensity, with $15,800 liquidated; at 4 hours, shorts continued to dominate, 4.16 times the longs, the squeeze intensity weakened but persisted, liquidations rose from $15,800 to $30,200; at 12 hours, shorts again crushed longs, 11.9 times the longs, squeeze intensity sharply increased again, liquidations soared to $143,500; at 24 hours, shorts still dominated, with $285,500 liquidated versus $29,500 for longs, shorts 9.68 times longs — the whale completed a comprehensive short squeeze on SNDK across all timeframes, all four time dimensions aligned, shorts continuously harvesting, total liquidations exceeded $310,000. This is a textbook one-sided short squeeze with shorts controlling the entire process. Everyone should manage positions carefully to avoid being harvested back and forth.
⚠️ Risk Warning: Shorts on all SNDK cycles continue to crush longs with highly consistent direction, but the 4-hour ratio narrowed to 4.16 before expanding again to 10-12 times at 12H/24H, showing some fluctuation in squeeze momentum; 24-hour liquidations account for 94% of the daily total, indicating high concentration. Leverage is recommended to be compressed below 3x, avoid blindly shorting, strictly control positions and wait for clearer direction.
🔥 Market Indicator | August 15
Today's three hot topics point to the same theme: the macro "stagflation" dilemma remains unresolved, but the AI sector has already entered a new phase of "heavy capital, high valuation, rapid capacity expansion."
📉 Weakening Consumer Momentum: No Rate Cuts, No Rate Hikes
US July retail sales fell 0.6% month-over-month, the largest drop in 14 months; University of Michigan consumer sentiment preliminary index dropped to 51, well below the expected 55. Consumer anxiety about the future is turning into actual spending contraction.
However, inflation stickiness still firmly limits policy space. July core CPI rose 2.5% year-over-year, exceeding the Fed's 2% target for the sixth consecutive year. CME data shows the probability of a rate hike in September has dropped to 28.6%, but this is not a "prelude to rate cuts," rather an awkward wait due to "lack of strength to hike." BMO analysts bluntly state retail data will "support the Fed maintaining rates next month" — no change, not because it's enough, but because they dare not move.
🤖 OpenAI vs. Anthropic Valuation Race: Bubble or Revolution?
The AI valuation race is heating up. OpenAI completed a $7 billion buyback at an $852 billion valuation, but executives are leaving one after another, and the gap between revenue and burn rate is tearing market confidence.
Meanwhile, Anthropic is expected to go public in October, with some investors valuing it as high as $2 trillion. This "sky-high" valuation is supported by its enterprise-grade large model API market share reaching 32%, surpassing OpenAI's 25%. By enterprise revenue metrics, annualized revenue is expected between $100 billion and $120 billion by the end of 2026.
A five-year-old company with a $2 trillion valuation. The market is betting not on profits but on AI's complete restructuring of the enterprise market.
🏗️ SK Hynix $720 Billion Capacity Expansion: Betting on AI Computing Power "Never Sleeping"
Storage leader SK Hynix announced a $720 billion investment to build the world's largest memory factory network, targeting HBM capacity expansion. The company clearly stated: memory has upgraded from a component to a core AI infrastructure.
Can returns be realized? In Q1, SK Hynix held 58% of the HBM market share, and order visibility seems stable. But the biggest risk is the mismatch between expansion cycles and demand fluctuations — its US stock has retraced about 21% from the July peak. If AI demand growth slows, the hundred-billion capacity could turn from a "moat" into a "cost black hole."
💎 Summary
Weakening consumption, persistent inflation — macro is hovering on the edge of "stagflation"; AI valuations from $852 billion to $2 trillion, the market is pricing next-gen enterprise technology with real money; SK Hynix bets $720 billion on AI computing power demand never fading. When macro weakness, high valuations, and heavy asset expansion act simultaneously — the AI sector is moving from "storytelling" to a "real money" critical test phase. #霍尔木兹协议待落地，原油风险等待定价
#标普盈利超预期，华尔街为何仅看7894点
#消费动能转弱，9月政策仍受通胀制约 US July retail sales decline, Fed expectations redefine BTC support level. The probability of a rate cut in September retreated from 50% to the 30% range, so why does the market interpret this positively for BTC? July US retail sales fell 0.6% month-over-month, significantly missing the market expectation of +0.1%. Excluding autos, sales also dropped 0.3%, and based on the control group, down 0.4%. While this does not indicate a consumption collapse, it clearly signals a momentum slowdown. During the same period, the probability of a September rate cut fell from around 50% a month ago to the low 30% range currently. This reflects the market pushing back the timing of the cut while repricing the cut path due to growth slowdown. - Key data: July retail sales -0.6% MoM, -0.7 percentage points versus expected +0.1% - Derivative market signal: September cut probability in the low 30% range, down from 50% a month ago - BTC spot price: about $63,000, with $62,000–$62,500 as key support levels - ETH spot price: about $1,880, relatively compared to BTC $BTC's four-year cycle pendulum is starting to swing again.
Brothers, if you still believe in this pattern, then you really need to keep your eyes wide open at this moment. Looking back at historical data, this rhythm is as precise as an alarm clock:
The bull market from 2015 to 2017 lasted exactly 1064 days; followed by the bear market from 2017 to 2018, exactly 364 days. Then the bull market from 2018 to 2021 was again exactly 1064 days; the bear market from 2021 to 2022 was still 364 days. Moving forward, the bull market from 2022 to 2025 is, no more no less, 1064 days.
This astonishing coincidence is hard to dismiss as mere coincidence. If history truly continues to follow the same rhythm, then the upcoming scenario is clear:
The bear market from 2025 to 2026 will most likely last 364 days. Counting on fingers, the potential major cycle bottom might fall around October 5, 2026.
Of course, history repeats but never simply duplicates. We can't be like the fool carving a mark on a boat to find a sword, but with this data in front of us, it's at least worth making a mental note. After all, in this circle, a bit more respect and a bit less recklessness is never a bad thing. Time will give the final answer; we'll watch and move forward.
#加密估值转向收入，BTC如何定价？ 📊 $ETH Liquidation Flash Report (August 15)
According to liquidation data, the whales have completed a comprehensive long squeeze on ETH across short to long timeframes, with longs being crushed from 1 hour friction to 24 hours, accumulating liquidations exceeding $1.49 million.
Time Total Liquidations Long Liquidations Short Liquidations
1 hour $294,000 $289,500 $4,447.95
4 hours $438,100 $407,800 $30,300
12 hours $778,400 $459,800 $318,600
24 hours $1,490,000 $813,200 $676,700
From the $ETH liquidation data, 1-hour long liquidations overwhelmingly surpass shorts, with longs 65 times the shorts, indicating a nuclear-level intensity long squeeze, with $294,000 liquidated; at 4 hours, longs continue to dominate, 13.5 times the shorts, the long squeeze intensity significantly weakens but remains strong, liquidations rising from $294,000 to $438,100; at 12 hours, longs still dominate, 1.44 times the shorts, long squeeze momentum sharply declines, liquidations moderately climb to $778,400; at 24 hours, longs continue to dominate, with $813,200 long liquidations versus $676,700 shorts, longs 1.2 times shorts—the whales have completed a comprehensive long squeeze on ETH across all timeframes, with consistent direction across four time dimensions, continuous long harvesting, and total liquidations exceeding $1.49 million. However, the key is the long dominance ratio shrinks from 65 times at 1 hour to 1.2 times at 24 hours, with long squeeze energy nearly exhausted, longs and shorts returning to balance, and direction possibly reversing at any time. Everyone should manage positions carefully to avoid being harvested back and forth.
⚠️ Risk Warning: All ETH timeframes show long liquidations continuously crushing shorts with highly consistent direction, but the ratio narrows from 65 times at 1 hour to 1.2 times at 24 hours, indicating a sharp decline in long squeeze momentum and a very high risk of direction reversal; 24-hour liquidations account for 93% of the daily total, showing extreme concentration and volatile market. Leverage is recommended to be compressed to within 3x, avoid blindly bottom-fishing, strictly control positions, and wait for clear direction.
🔥 Market Indicator | August 15
Today's three hot topics point to the same theme: the macro "stagflation" dilemma remains unresolved, but the AI sector has already entered a new phase of "heavy capital, high valuation, rapid capacity expansion."
📉 Weakening Consumer Momentum: No hope for rate cuts, no courage for hikes
US July retail sales fell 0.6% month-on-month, the largest drop in 14 months; University of Michigan consumer sentiment preliminary index dropped to 51, well below the expected 55. Consumer anxiety about the future is turning into actual spending contraction.
But inflation stickiness still firmly locks policy space. July core CPI rose 2.5% year-on-year, exceeding the Fed's 2% target for the sixth consecutive year. CME data shows the probability of a rate hike in September has dropped to 28.6%, but this is not a "prelude to rate cuts," rather an awkward wait due to "lack of strength to hike." BMO analysts bluntly state retail data will "support the Fed maintaining rates next month"—no change, not because it's enough, but because they dare not move.
🤖 OpenAI and Anthropic Valuation Race: Bubble or Revolution?
The AI valuation race has heated up. OpenAI completed a $7 billion buyback at an $852 billion valuation, but executives are leaving one after another, and the gap between revenue and burn rate is tearing market confidence.
Meanwhile, Anthropic is expected to go public in October, with some investors valuing it as high as $2 trillion. Supporting this "sky-high" valuation is its enterprise-grade large model API market share reaching 32%, surpassing OpenAI's 25%. By enterprise revenue metrics, 2026 annualized revenue is expected between $100 billion and $120 billion.
A five-year-old company with a $2 trillion valuation. The market is betting not on profits but on AI's complete restructuring of the enterprise market.
🏗️ SK Hynix $720 Billion Capacity Expansion: Betting on AI Computing Power "Never Sleeping"
Storage leader SK Hynix announced a $720 billion investment to build the world's largest memory factory network, targeting HBM capacity expansion. The company clearly states: memory has upgraded from a component to a core AI infrastructure.
Can returns be realized? In Q1, SK Hynix held 58% of the HBM market share, and order visibility seems stable. But the biggest risk is the mismatch between expansion cycles and demand fluctuations—its US stock has retraced about 21% from the July peak. If AI demand growth slows, the hundred-billion capacity could turn from a "moat" into a "cost black hole."
💎 Summary
Consumer weakness and persistent inflation—macro is hovering on the edge of "stagflation"; AI valuations from $852 billion to $2 trillion, the market is pricing next-generation enterprise technology with real money; SK Hynix bets on AI computing power demand never fading with a $720 billion expansion plan. When macro weakness, high valuations, and heavy asset expansion act simultaneously—the AI sector is moving from "storytelling" to a "real money" critical test phase. #消费动能转弱，9月政策仍受通胀制约
#OpenAI与Anthropic估值竞赛升温
#海力士扩产提速，资本开支能否兑现回报 If you look only at macro logic, the current market seems more optimistic. With Fed rate cut expectations heating up and global liquidity marginally improving, the biggest beneficiaries should be crypto, a high-beta risk asset. But the answer given by the market isn't 'chickens and dogs rising to the sky,' but rather an increasingly obvious divergence between strength and weakness. $BTC No new main rally has been triggered by rate cut expectations, and most altcoins have not seen sustained broad-based gains. This actually shows a problem: the market isn't out of money, but rather that money has become more selective. In past liquidity markets, capital often spread layer by layer from $BTC → $ETH → mainstream altcoins → small-cap → memes; But now, it's more like capital is first concentrated in a few certain assets, then seeking local high-odds opportunities. So at this stage, what matters more than predicting the "next tenfold coin" is understanding who the capital is giving a valuation premium. First Tier: $BTC Remains the liquidity anchor for the entire market$BTC remains the core indicator to watch this round of the market. Short-term price fluctuations are not scary; what really matters is whether new funds can form continuous inflows again. ETFs, institutional allocation, and macro liquidity expectations determine the $BTC's ability to support the downside; But without new incremental buying, relying solely on expectations of "future rate cuts" will make it difficult to keep pushing prices upward. In other words, rate cut expectations are the background, not the buying momentum itself. What the market truly lacks now is a clear capital confirmation. As long as $BTSideways movement doesn't mean there's no story; the story is waiting for a bullish candlestick
BTC has been grinding within the 63,000-65,000 range for nearly ten weeks.
On the surface, it looks like liquidity is drying up, but looking deeper, four types of forces are simultaneously converging.
Price momentum hits bottom first. Both daily and weekly RSI show bullish divergence—price hasn't broken the previous low, while the indicator's bottom is rising, a classic sign that bears are losing strength.
Volatility collapses first. Bollinger Band width has narrowed to the tightest since January, BTC DVOL dropped to around 35%. Historically, this kind of "silence" lasts 3-5 weeks on average before a directional move, and it will be a volume-backed breakout.
Chips are quietly being accumulated. Last week, BTC+ETH spot ETFs had a combined net inflow of $1.1 billion, with IBIT alone taking about $693 million, accounting for 80%. In the same week, on-chain exchanges saw a net outflow of 12,400 BTC, while LTH supply increased by 41,000 BTC—institutions are sweeping on the ETF side, long-term holders are locking on-chain, and floating chips are being squeezed from both ends.
Leverage dies first. Funding rates returned to around 0.01%, three times below the "overheat" threshold of 0.05%. High-leverage longs have been liquidated, cleaning up the market.
Adding the macro backdrop of July CPI year-over-year at 3.4% and core CPI at 2.5%, continuing downward, tightening pricing is retreating, and the opportunity cost of risk assets is decreasing.
Technicals are holding back, funds are flowing in, on-chain is locking up, macro is easing—four conditions are met simultaneously, all that's missing is a volume-backed weekly close. Following Bitcoin’s lead, where does Ethereum go from here? 🔥 Consumer momentum is weakening, and September policy is still constrained by inflation. Most traders still think in single-asset inertia: BTC drags the whole crypto market up and down. But over the next phase, the divergence between BTC and ETH will widen, and their correlation will fade in stages. That is the biggest trap in this market right now. Starting with $BTC: Bitcoin’s valuation logic is gradually decoupling from crypto-nati#OpenAI与Anthropic估值竞赛升温
1. Real-time core data
Anthropic's latest Series H financing valuation is $965 billion, surpassing OpenAI's $852 billion; the former's Q2 annualized revenue is $47 billion, with growth far exceeding its competitor. Institutions predict an IPO valuation impact of $2 trillion in October.
Capital flow overview: BTC current price 62860, spot ETF has seen net outflows for three consecutive days; AI sector RNDR current price 3.85 (24h +0.7%), FET current price $0.51 (24h +1.9%), funds are beginning to shift from small AI coins without landing to leading computing power players.
2. Underlying logic
The valuations of the two major AI giants are intensifying competition, with hundreds of billions in financing continuously absorbing market liquidity. US AI equities have become the preferred allocation for funds, diverting hot money originally invested in crypto AI themes.
Sector differentiation logic: RNDR is tied to GPU computing power, aligning with the essential AI training needs of large companies, with a solid long-term narrative; FET focuses on AI agents, fitting enterprise automation demands; in contrast, small-cap AI coins like $ROBO rely solely on thematic speculation, facing greater capital outflow pressure.
Mid-to-long-term positives: Global AI infrastructure investment is increasing, and the decentralized computing power sector has ongoing fundamental support, though short-term gains are suppressed by massive primary market financing.
3. Personal view
Operate cautiously, avoid speculative small-cap AI coins without practical application; prioritize buying RNDR and FET, which have mature fundamentals, and do not chase short-term pulse rallies. The short-term siphoning effect of the primary market will continue to suppress altcoins; increase positions after the Anthropic IPO lands and capital flow pressure eases.
This is only a personal opinion and does not constitute investment advice #消费动能转弱，9月政策仍受通胀制约 #OpenAI与Anthropic估值竞赛升温 The "wealth relay baton" of the rate cut cycle: BTC is the prelude, ETH is the climax
Core logic (must read):
1. Rate cut trades proceed in three steps: Certainty (BTC) → Elasticity (ETH) → Sentiment (Altcoins). We are currently in the critical transition from the first to the second step.
2. Key indicator: Don’t trust slogans, trust exchange rates. If ETH/BTC doesn’t rise, firmly don’t talk about the "altcoin season." This is the watershed between a "rebound" and a "reversal."
3. Fatal risk: Beware of a "recession-style rate cut"! If employment data collapses, BTC will initially rise due to liquidity expectations but will then plummet due to risk-off sentiment. ETH will be the hardest hit.
4. Operational strategy:
· Current phase: Watch BTC to trade ETH. If BTC is stable, then ETH moves in.
· Breakout signal: ETH/BTC strengthens for three consecutive days; this is the "starting gun" for market risk appetite expansion.
· Retreat signal: Economic hard landing data is released; any rebound is a bull trap.
Remember: The real big market move is not when BTC rises the most fiercely, but when the market starts to find BTC "rising too slowly." At that time, ETH’s scythe and the frenzy will arrive simultaneously.$CORE market shows slight recovery, and many loyal fans start subjective speculation again, constantly imagining various major positive catalysts. Many people assume that long-term benefits will definitely materialize, firmly believing that institutional funds will soon enter the market massively, prematurely overdrawing valuations for the 2027 and 2028 bull markets. Objectively speaking, to trigger a trend-level major rally, the conditions are extremely stringent: smooth approval of compliantThe Wrong Safe Haven?
Trump brings Iran back into focus, and markets instantly turn defensive.
Gold rises.
Treasuries gain.
The dollar strengthens.
BTC gets hit.
That’s the key: when fear spikes, institutions usually cut high-volatility assets first. BTC may be called “digital gold,” but in a real risk-off move, it still trades more like a risk asset.
Don’t confuse BTC with a safe haven just yet. In panic, it may be the first thing sold—not the first thing bought.
$BTC $ETH
#DailyOrbit 📊 $DOGE Contract Liquidation Express (August 15)
According to liquidation data, the dog whales have completed a comprehensive long liquidation from short to long cycles on DOGE. Shorts were continuously wiped out in the short cycle, and longs didn’t even get a moment to breathe...
Time Total Liquidation Long Liquidation Short Liquidation
1 hour $42,700 $42,700 $0
4 hours $146,800 $146,700 $89.89
12 hours $155,400 $155,300 $89.89
24 hours $235,600 $195,700 $39,900
From the $DOGE liquidation data, in the 1-hour period, long liquidations crushed shorts, with shorts completely wiped out. The long liquidation event unfolded with nuclear-level intensity, with $42,700 liquidated; at 4 hours, longs continued to crush shorts, with longs 1,632 times the shorts, maintaining extremely high liquidation intensity, with liquidation volume jumping from $42,700 to $146,800; at 12 hours, longs still dominated, with longs 1,728 times the shorts, and liquidation intensity continued to rise moderately to $155,400; at 24 hours, longs still dominated, with $195,700 long liquidations versus $39,900 shorts, longs 4.9 times the shorts—dog whales completed a comprehensive long liquidation from short to long cycles on DOGE. Shorts were completely wiped out in 1 hour, and although present in mid-to-long cycles, their strength was extremely weak, with cumulative liquidations exceeding $235,600. This is a textbook-level one-sided long liquidation event. Everyone, control your positions well and avoid being harvested back and forth.
⚠️ Risk Warning: Long liquidations on DOGE continue to crush shorts across all cycles, with highly consistent direction, but the 24-hour multiple suddenly dropped to 4.9 times, indicating a sharp decline in long liquidation momentum. Beware of the risk of a directional reversal; 24-hour liquidation volume accounts for 90% of the total daily volume, showing extremely high concentration. Leverage is recommended to be compressed to within 3x; do not blindly bottom-fish, strictly control positions and wait for clear direction.
🔥 Market Indicator | August 15
Today’s three hot topics point to the same theme: the macro "stagflation" dilemma remains unresolved, but the AI sector has already entered a new phase of "heavy capital, high valuation, rapid capacity expansion."
📉 Weakening Consumer Momentum: No hope for rate cuts, no courage for hikes
U.S. retail sales in July fell 0.6% month-over-month, the largest drop in 14 months; the University of Michigan Consumer Sentiment Index preliminary reading dropped to 51, well below the expected 55. Consumer anxiety about the future is turning into actual spending contraction.
However, inflation stickiness still firmly locks policy space. July core CPI rose 2.5% year-over-year, exceeding the Fed’s 2% target for the sixth consecutive year. CME data shows the probability of a rate hike in September has dropped to 28.6%, but this is not a "prelude to rate cuts," rather an awkward wait due to "lack of strength to hike." BMO analysts bluntly state that retail data will "support the Fed in keeping rates unchanged next month"—no move, not because it’s enough, but because they dare not move.
🤖 OpenAI and Anthropic Valuation Race: Bubble or Revolution?
The AI valuation race has entered a white-hot stage. OpenAI completed a $7 billion buyback at an $852 billion valuation, but executives are leaving one after another, and the gap between revenue and burn rate is tearing market confidence.
Meanwhile, Anthropic is expected to go public in October, with some investors valuing it as high as $2 trillion. Supporting this "sky-high" valuation is its enterprise-level large model API market share reaching 32%, surpassing OpenAI’s 25%. According to enterprise revenue statistics, annualized revenue by the end of 2026 is expected between $100 billion and $120 billion.
A five-year-old company with a $2 trillion valuation. The market is betting not on profits but on AI’s complete restructuring of the enterprise market.
🏗️ SK Hynix $720 Billion Capacity Expansion: Betting on AI Computing Power "Never Sleeping"
Storage leader SK Hynix announced a $720 billion investment to build the world’s largest memory factory network, targeting HBM capacity expansion. The company clearly stated: memory has upgraded from a component to a core AI infrastructure.
Can returns be realized? In Q1, SK Hynix held 58% of the HBM market share, and order visibility seems stable. But the biggest risk is the mismatch between expansion cycles and demand fluctuations—its U.S. stock has retraced about 21% from the July peak. If AI demand growth slows, the hundred-billion capacity could turn from a "moat" into a "cost black hole."
💎 Summary
Consumer weakness and persistent inflation—macro is hovering on the edge of "stagflation"; AI valuations from $852 billion to $2 trillion, the market is pricing next-generation enterprise technology with real money; SK Hynix is betting on AI computing demand never fading with a $720 billion expansion plan. When macro weakness, high valuations, and heavy asset expansion act simultaneously—the AI sector is moving from "storytelling" to a "real money" critical test phase. #消费动能转弱，9月政策仍受通胀制约
#OpenAI与Anthropic估值竞赛升温
#海力士扩产提速，资本开支能否兑现回报 Anthropic > OpenAI.
Here's the actual case.
Everyone's watching the valuation race, but the real signal is revenue quality, not size.
Anthropic went from $1B to $47B+ ARR in about 16 months, the fastest compounding enterprise software company in history. And it's not hype-driven consumer traffic, it's enterprise: 8 of the Fortune 10 as customers, 500+ accounts paying $1M+. Claude Code alone is running a $2.5B ARR.
OpenAI has scale. Anthropic has margin discipline targeting 77% gross margins by 2028, built on actual paying enterprise workloads instead of subsidized consumer usage.
When the AI-stock correction hits (and it will), the companies with real enterprise revenue survive it. The ones running on consumer hype don't.
Which one do you think holds up better when the froth clears?
#OpenAIAnthropicRace The market has been showing a very contradictory vibe these past two days. On one hand, $BTC is still fluctuating around $63,000 without a true breakout; on the other hand, a large number of altcoins have clearly weakened, especially those previously boosted by sentiment and trading volume. Once funds retreat, the pullback speed is faster than during the rise. This actually illustrates a problem: the market is not lacking funds, but rather that capital is becoming increasingly selective. Let's start with $LAB. This coin has fallen from around 0.65 all the way to now, with the price returning to around $0.08, a retracement of over 90% from its previous high. Looking only at the decline, it's easy to get the illusion that "maybe it's almost already dropped," but this mindset is often the most dangerous in high-volatility, small-cap assets. Because a 90% drop doesn't mean the remaining 10% is safe. $LAB There is still potential unlocking pressure ahead, and currently, the market's tolerance for projects with small market caps, high FDV, and continuous chip release is clearly declining. Especially when there is no new capital continuously taking over, unlocking is not just a fundamental event; it can easily turn into a liquidity shock on the market floor. So for this kind of coin, I now prefer to think of it as a "trading target" rather than a "sufficiently cheap allocation asset." What truly matters is where mainstream capital is headed. $BTC Currently still fluctuating repeatedly around $63,000. After the initial push to $65,000, it didn't hold immediately, indicating selling pressure still exists above, but one thing is clear: funds haven't existedIs the big one coming? Is the big one coming?
Do I, the mold guy, still have to get up today? Yesterday was the APR crash, and today it's $CAP's turn?
Altcoins have been falling these past two days, APR plummeted in minutes yesterday, which really boosted my confidence.
Lying in my rental, I feel the future is bright; I haven't even digested yesterday's big meal, and today it feels like there's more meat to eat.
You might call me crazy, nuts screwed up? The trend is so strong, soon it'll explode on you.
Don't rush. Haven't you noticed the funding rate has changed—from -0.9% to -0.2%, the shorts' incentive to pay is weakening, and longs are increasing.
More importantly, the highs keep dropping—from 0.078 down to 0.07, then to 0.062; each rebound is weaker, and the strength of buyers taking over is diminishing.
The long-short ratio tells more—the Binance large accounts' long-short ratio is as high as 1.6556, longs are extremely crowded.
Such a severely overcrowded structure, once profits are taken, the sell-off will be very rapid.
All three signals are lit. I'm holding short positions, stop loss at 0.066, target first at 0.055, and if broken, look at 0.04.
Yesterday's big meal with egg and chicken leg, today I'll try to add a bottle of drink.
$BTC #消费动能转弱，9月政策仍受通胀制约
$APR BTC: The Biggest Bounce May Be Coming Look at the symmetry of $BTC . As a bull market matures, corrections tend to become deeper. In a maturing bear market, the opposite can happen: the broader trend remains bearish, but rebounds become steeper and increasingly aggressive. If $BTC sees another sharp decline, sentiment could quickly shift toward expectations of even lower prices. Yet that extreme bearishness may be exactly what sets the stage for the strongest rebound of the entire bear market.#WYour sense of this market is pretty good, catching the rotation rhythm quite accurately. AI and storage just took a breather, and capital always needs an outlet. The power sector line is indeed worth pondering—when computing power stacks up, it all comes down to the electricity bill, and the market will have to reckon with this sooner or later.
Back to these two Pre-IPO AI targets. Their price movements are almost cut from the same mold, indicating this wave is driven by emotional resonance rather than significant differences in their fundamentals. But the fundamentals themselves are worth breaking down:
· Anthropic hit 11.5 billion in revenue in Q2, more than doubling quarter-over-quarter, and has turned profitable, essentially "proving it can make money."
· OpenAI’s annualized revenue has surged to 40 billion, also growing fiercely, but profitability is expected only by 2030, following a "capture market first, profit later" strategy.
The valuation gap is nearly double, and the market has already shown its stance—now it values profitability over pure storytelling. The AI sector is shifting from "imagination premium" to "performance validation."
Everyone knows the advantage of Pre-IPO: getting in early without waiting for the public listing. The problem is that a lot of expectations are already priced in, so the remaining upside depends on beating those expectations, which is not easy. Plus, these contract types have limited liquidity, making volatility easy to get carried away with. It looks exciting, but if you really get in, you need to consider your risk tolerance carefully. Better to wait and not rush.
Storage, AI, power—the industrial chain is tightly linked. If capital really rotates, power will be remembered sooner or later. As for when, there’s no use rushing; just wait.
$SNDK
#消费动能转弱，9月政策仍受通胀制约
#OpenAI与Anthropic估值竞赛升温
#海力士扩产提速，资本开支能否兑现回报 The recent round of the crypto market doesn't seem lively on the surface. $BTC has been fluctuating repeatedly around $63,000, altcoins haven't shown the collective valuation surges seen in the past, and trading volume is far from enthusiastic. If you only focus on price changes, it's easy to conclude: the market has no opportunities, and funds seem to be withdrawing. But if you break down the flow of funds, you'll find the opposite is true. Money hasn't completely left the crypto market; it's just no longer willing to pay for all assets together. This is the most noteworthy change in the market recently, in my opinion. In past typical risk-on cycles, $BTC usually rises first, followed by $ETH, then large-cap altcoins and popular sectors, and finally funds spread to various small-cap projects. As long as liquidity is sufficient, the core of market trading isn't asset quality but "who hasn't risen yet." Now, this logic is clearly weakening. Funds are switching from "buying sectors" to "picking assets." According to current data, the total crypto market cap remains relatively high, with $BTC dominance around 56.9%, and the Fear & Greed Index near 62. This combination is interesting: sentiment isn't fearful, but it's far from the stage of recklessly chasing risk assets regardless of price. In other words, the market has risk appetite, but not an excessive one. This directly determines where funds will flow next. Currently, I prefer to divide market funds into several lines. First are still $BTC and $ETH. $BTC is fluctuating around $63,000, and recently spotFourth Halving, Day 846: BTC almost zero returns.
It's not a cycle failure, it's a cycle compression.
① Peak gains: +8885% → +2752% → +560% → +102%
② Same-day comparison: 23x → 9.8x → 2.1x → 1.1x
③ Drawdown -49% / 312 days (historical bear markets were deeper)
④ Mayer 0.74 · Pi 0.70 · FNG 34
ETF era: the time pattern remains, amplitude is narrowing.
All four cycles overlapped—do you think the compression will continue? $BTC $ETH $OKB 📊 $BTC Contract Liquidation Express (August 15)
According to liquidation data, whales played a three-stage harvesting strategy on BTC: short-term long liquidation → mid-term short squeeze → long-term long liquidation again, repeatedly switching directions and harvesting both longs and shorts, with total liquidations exceeding $1.98 million.
Time Total Liquidation Long Liquidation Short Liquidation
1 hour $11,700 $9,974.99 $1,703.26
4 hours $61,800 $40,700 $21,000
12 hours $476,600 $135,700 $340,900
24 hours $1,986,400 $1,068,600 $917,700
From $BTC liquidation data, in 1 hour long liquidations crushed shorts, longs were 5.86 times shorts, with a nuclear-level intensity long liquidation totaling $11,700; at 4 hours longs continued to dominate, longs were 1.94 times shorts, long liquidation intensity sharply weakened, total liquidations rose from $11,700 to $61,800; at 12 hours the direction completely reversed, short liquidations crushed longs, shorts were 2.51 times longs, whales completed a fierce turn from long liquidation to short squeeze, with liquidations soaring to $476,600; at 24 hours the direction reversed again, long liquidations at $1,068,600 versus shorts at $917,700, longs were 1.16 times shorts—whales completed the three-stage harvesting of long liquidation → short squeeze → long liquidation again on BTC, repeatedly switching directions and harvesting both sides, with total liquidations exceeding $1.98 million. But importantly, the 24-hour long-short ratio was only 1.16, long liquidation energy was nearly exhausted, longs and shorts returned to balance, and direction could reverse again at any time. Everyone should control positions carefully and avoid being harvested back and forth.
⚠️ Risk Warning: BTC multi-timeframe directions repeatedly switch (1H long liquidation → 4H long liquidation weakening → 12H short squeeze → 24H long liquidation), with extremely intense direction changes; 24-hour long-short ratio only 1.16, though direction favors long liquidation, the strength is very weak, beware of further repeated risks; 12-hour + 24-hour liquidations account for 99% of the daily total, with very high concentration and volatile market. Leverage is recommended to be compressed to within 3x, avoid chasing highs or selling lows, strictly control positions and wait for clear direction.
🔥 Market Indicator | August 15
Today's three hot topics point to the same theme: the macro "stagflation" dilemma remains unresolved, but the AI sector has already entered a new phase of "heavy capital, high valuation, rapid expansion."
📉 Consumer Momentum Weakens: No Hope for Rate Cuts, No Courage for Rate Hikes
US July retail sales fell 0.6% month-on-month, the largest drop in 14 months; University of Michigan consumer sentiment index preliminary reading dropped to 51, well below the expected 55. Consumer anxiety about the future is turning into actual spending contraction.
But inflation stickiness still firmly locks policy space. July core CPI rose 2.5% year-on-year, exceeding the Fed's 2% target for the sixth consecutive year. CME data shows the probability of a rate hike in September has dropped to 28.6%, but this is not a "prelude to rate cuts," rather an awkward wait due to "lack of strength to hike." BMO analysts bluntly say retail data will "support the Fed maintaining rates next month"—no change, not because it's enough, but because they dare not move.
🤖 OpenAI and Anthropic Valuation Race: Bubble or Revolution?
The AI valuation race has heated up. OpenAI completed a $7 billion buyback at an $852 billion valuation, but executives are leaving one after another, and the gap between revenue and burn rate is tearing market confidence.
Meanwhile, Anthropic is expected to go public in October, with some investors valuing it as high as $2 trillion. Supporting this "sky-high" valuation is its enterprise-level large model API market share reaching 32%, surpassing OpenAI's 25%. By enterprise revenue metrics, annualized revenue by the end of 2026 is expected between $100 billion and $120 billion.
A five-year-old company with a $2 trillion valuation. The market is betting not on profits but on AI's complete restructuring of the enterprise market.
🏗️ SK Hynix $720 Billion Expansion: Betting on AI Computing Power "Never Sleeps"
Storage leader SK Hynix announced a $720 billion investment to build the world's largest memory factory network, targeting HBM capacity expansion. The company clearly stated: memory has upgraded from a component to a core AI infrastructure.
Can returns be realized? In Q1, SK Hynix held 58% of the HBM market share, and order visibility seems stable. But the biggest risk is the mismatch between expansion cycles and demand fluctuations—its US stock has retraced about 21% from the July peak. If AI demand growth slows, the hundred-billion capacity could turn from a "moat" into a "cost black hole."
💎 Summary
Consumer weakness and persistent inflation—macro hovers on the edge of "stagflation"; AI valuations from $852 billion to $2 trillion, the market is pricing next-generation enterprise technology with real money; SK Hynix bets on AI computing power demand never fading with a $720 billion expansion plan. When macro weakness, high valuations, and heavy asset expansion act simultaneously—the AI sector is moving from "storytelling" to a "real money" critical test phase. #消费动能转弱，9月政策仍受通胀制约
#OpenAI与Anthropic估值竞赛升温
#海力士扩产提速，资本开支能否兑现回报 LTH Profit Supply Share (LTH Supply In Profit Share) measures: the proportion of BTC held by long-term holders that is currently still in profit.
(1) Very high value (close to 100%): Often appears near the late bull market or top, where long-term holders generally have significant profits and may start distributing (selling).
(2) Lower value (close to 50%): Often appears in the mid to late bear market, where a large proportion of long-term holders are at a loss, historically corresponding to accumulation phases or potential bottom areas.
(3) This cycle: Near the BTC price low on 2026-06-30: a minimum of about 54.64%, with on-chain structure already showing characteristics of a cycle bottom. However, according to historical cycle patterns, there is still about a 2-month window from the high to the bear market bottom for BTC, which is what makes this cycle different from previous ones.$BTC should completely turn bearish, how many friends have realized this: 1. Both CPI and PPI data point to no rate hikes, so logically BTC should rise slightly, but instead ETFs have seen net outflows for two consecutive days. Fidelity and ARKB are leading redemptions, even IBIT, the strongest buyer, can't help but withdraw. Even positive expectations can't drive the trend, which shows how weak it is. 2. This is not exactly a case of "positive news turning into negative." Because positive newsOn August 13, Tether announced that KPMG USA had issued an "unqualified opinion" on Tether International's annual financial statements for the year ended December 31, 2025. Let's clarify the timeline: this is the result announced in August 2026, but the audit target is fiscal year 2025, not a real-time proof of today's reserve status. The difference between this and previous quarters' reserve assurance lies in the more complete audit scope. According to Tether's announcement, KPMG reviewed its balance sheet, income statement, equity changes, and cash flows, and tested transactions, systems, ownership records, valuations, and counterparty evidence; Gold reserves also underwent physical inventory. Audited reports show reserves exceeding liabilities by $6.814 billion. For comparison, Tether's Q4 2025 assurance, released in January this year and completed by BDO, shows total assets of about $192.878 billion, total liabilities of about $186.54 billion, and excess reserves of about $6.3 billion. The two sets of numerical calibers may differ from audit adjustments and cannot be mechanically subtracted. Why is this matter worth the attention of ordinary crypto users? USDT is one of the most widely used stablecoins for trading, settlement, and hedge swaps. If full annual audits can continue to be disclosed, it could improve the efficiency of institutions and regulators in assessing reserve quality, liquidity, and internal controls, and also raise the transparency threshold for other stablecoin issuers. However, an "unqualified opinion" only indicates that the financial statements materially comply with applicable accounting standards and does not mean that the underlying assets have no price or liquidity riskYesterday's overall market was still in a state of "BTC moving sideways, with funds continuing sector rotation." 1. Yesterday's Market: $BTC No Directional Move, Funds Continue to Rotate BTC Yesterday still failed to break a clear trend, with prices continuing to fluctuate at key levels. However, internal rotation within the altcoins is quite obvious: earlier, OKB was the first to strengthen, and now $LINK and the oracle sector are taking over. This indicates that on-market funds have not completely withdrawn; for now, there has been no BTC-led rally with mainstream coins following the rally; it seems more like funds are seeking opportunities across different sectors. 2. Technical Perspective: BTC Weekly Chart Has Reached a Key Level Right now, my main focus is still on BTC Weekly Chart. The price has just reached near the major downtrend line. At this level, you can't just look at "breaking through"—the truly effective move should be: break the trendline→ pullback → pullback if it doesn't fall back, → hold steady again. Only if this process truly emerges will I believe BTC's mid-term structure is starting to strengthen. If the breakout immediately falls back below the trendline, then continue to guard against a false breakout. 3. Macroeconomic Side: The US Dollar is Weakening, but BTC Has Not Fully Absorbed the Positive Factors On the US side, there has actually been a relatively friendly shift for risk assets. On August 14, the US dollar index fell to around 99.65, down about 0.28% in a single day. The main reason is that U.S. retail sales unexpectedly fell 0.6% in July, and combined with previous moderate inflation data, market expectations for continued Fed rate hikes in September have clearly declined. So from a macro perspective,[Major Underlying Crisis! Ethereum and SOL Simultaneously Fall into a Merton Dilemma]
Since entering the industry in 2016, I have always emphasized that the long-term value of public blockchains is never determined by short-term market speculation, but by whether the economic model can sustain operation. Recently, Ethereum and Solana have both exposed irreconcilable structural contradictions, which all holders of ETH, SOL, and LST assets must take seriously.
The Ethereum EIP-8363 proposal plans to reduce validator staking rewards. Once implemented, staking yields could be directly halved, dropping from 2.86% to 1.476%. The proposal was immediately met with strong resistance from the community. Currently, 41.4 million ETH are staked, amounting to $79.6 billion, supporting a $35 billion LST derivative lending and leverage system. Once staking rewards are reduced, interest rate sectors like Pendle will be repriced, and the entire DeFi benchmark interest rate system will face shocks.
On the other hand, Solana is also struggling. The on-chain inflation reform proposal is advancing, and the number of active validators has plummeted from a peak of 2,500 to 683. Although there are plans to increase SOL burn amounts, the daily burn volume is far from covering the issuance rewards, causing non-staking holders to be continuously diluted by inflation.
Two top public chains face the same dilemma: staking rewards are caught in a bind.
If high issuance rewards are maintained, token holders suffer continuous inflation dilution; if rewards are cut, validator earnings decline, leading to network exits and increased centralization risk. This is a classic Merton dilemma—no matter how policies adjust, new risks will emerge.
Many retail investors only focus on price fluctuations and ignore changes in underlying rules. LST staking, on-chain leverage, and interest rate derivatives are all built on stable staking yields. If the underlying economic model loosens, the derivative market is poised for a valuation reset at any time.
Everyone tends to be blindly optimistic in a bull market, but after experiencing multiple bull and bear cycles, I know that real risks often hide in the underlying mechanisms that everyone overlooks. Don’t just bet on market trends; closely follow public chain governance proposals to proactively avoid potential valuation shocks.
Do you think ETH and SOL will ultimately choose to sacrifice token holders or validators? Discuss in the comments.
#以太坊草案EIP-8363引争议 $ETH $SOL Newly entered trading funds remain in a unidirectional market, and traditional on-chain derivative pools have not seen the expected overflow liquidity.
About 169,000 wallets chose to trade in the RWA market for the first time in the first six months of this year, accounting for 31.7% of all new users.
Approximately 80.9% of retained RWA traders continue to trade this category without entering the crypto asset market to participate in spot or contract trading.
The category preference of incremental users directly cuts off the path of funds penetrating the entire site's liquidity pools, forming mutually isolated liquidity patterns within the platform.
If this retained capital later begins to penetrate crypto asset derivatives, and the isolation ratio gradually decreases from 80.9%, it will boost the overall depth and turnover efficiency of the entire site; this path fails if the isolation ratio continues to rise.
If the 31.7% growth rate of new users slows and RWA funds experience unidirectional withdrawal, the platform will face liquidity pressure due to interruption of incremental replenishment; this weakening logic fails when there is an independent large net inflow in the native crypto sector.
The current judgment on full-ecosystem liquidity coordination will be completely falsified when RWA funds and crypto assets exhibit high-frequency bidirectional transfers.
The most critical observation variable in the next 7 days is the actual proportion of RWA first-time users transferring funds to the crypto asset sector during subsequent position adjustments.
#交易之声：你的经验值得被听到 #闪迪投资者日后股价大涨，长期目标待验证 #英伟达深入AI资本链，协同与风险如何平衡亿万富翁马克·库班再次抛出了一个引发讨论的观点：计算机芯片将成为新的加密货币。 这句话是什么意思？ 库班的表述指向一个趋势：算力正在成为比代币本身更稀缺的资源。 在加密货币的早期阶段，代币（如BTC、ETH）是价值的直接载体。但随着AI和加密生态的融合，算力本身正在成为一种独立的资产类别——谁拥有芯片，谁就掌握了生产数字价值的能力。 代币 vs 芯片：两种“数字稀缺性”的底层逻辑 马克·库班说“芯片是新的加密货币”，但两者在本质上其实走的是两条不同的稀缺性路径： 供应机制不同：加密货币的供应由代码写死（如比特币2100万枚上限），不随需求变化；芯片的供应则取决于制造产能，虽可扩产，但建厂周期长、技术门槛高，短期弹性极低。 用途定位不同：加密货币的核心功能是价值存储与交易媒介，持有本身就是目的；芯片则是算力的生产工具，价值在于它能“造”出什么。 稀缺性来源不同：加密货币的稀缺来自数学上限；芯片的稀缺来自制造能力与技术壁垒的叠加——即便想造，也不一定造得出来。 价值捕获方式不同：持有加密货币，是持有资产本身；持有芯片，则是持有生产工具，需要通过运行算力来持续产生收益。 芯片作为硬资产，其价#dusk $DUSK @Dusk_Foundation Suppose I create a regulated asset on Dusk ($DUSK) and then move it to another public chain. What worries me the most isn’t actually how fast the transfer is. The real headache is: what exactly does the “entity” I’m moving over represent?
If it were just an ordinary token, this would be a purely technical matter—lock it here, mint it there, done. But when Dusk deals with RWA, their approach is clearly more complex. Their underlying logic is full of rules: who is qualified to buy, who can transfer, which core data must be kept confidential, and what must be disclosed under certain audits. If after cross-chain transfer, all these original holding thresholds and compliance restrictions are left behind at home, then what the other network sees at best is just a symbol disguised as a token, no longer possessing the full integrity of a traditional financial instrument.
This explains why Dusk decided to use Chainlink CCIP for the underlying cross-chain interaction, so that tokenized assets in DuskEVM can naturally move to other ecosystems to engage in financial activities. Honestly, this move completely reshaped my understanding of the term “cross-chain.”
Previously, people thought asset cross-chain transfers were mainly about gaining liquidity elsewhere. Looking back at the RWA business, the real challenge is how to ensure “the rules flow with the funds.” Think about it: if a security completes a cross-network transfer but the original investors’ eligibility, trading bans, and other constraints don’t follow, then what’s displayed on the new public chain is just an incomplete token, not the legally protected asset it once was.
So now I’m closely watching a key point after Dusk’s cross-network transfer: after the asset relocates, how much of its originally bound identity information, operational permissions, and regulatory clauses can be restored in the new environment? When doing RWA cross-chain, the real challenge is never about how far the funds are thrown, but ensuring that the accompanying compliance code of law can pack up and move intact along with them. #eth $ETHWhen the Frenzy Narrows: The Silent Cracks Behind the New Highs in U.S. Stocks
$SNDK $BTC In recent days, news about new highs in U.S. stocks has exploded like festive fireworks. The numbers for the S&P 500 and Nasdaq have been climbing steadily, but if you peel back the numerical shell, you'll find a disturbing fact — this rally feels like a party for a very few, with the familiar faces leading the dance: giant tech stocks. The narrow breadth of the rally has led many analysts to recall the tail end of the internet bubble.
Meanwhile, Wall Street quietly recalls the dark memory of 1987. On October 19 of that year, the Dow Jones dropped 22.6% in a single day, a shadow that still lingers in the minds of many veteran traders. Today's market is dominated by algorithmic trading, accounting for over 70%, with abnormally low volatility, yet the implied volatility in the futures market has already shown warning signs. A key question emerges: Is the productivity revolution brought by AI a genuine new engine, or just another narrative-blown illusion?
If it’s the latter, then the current valuations are like castles suspended in midair.
Two Faces at the Gambling Table
Interestingly, different social classes are making the same move at this moment — betting. The wealthy are betting on the housing market bottoming out. Over the past year, the median price of existing homes in the U.S. rose 4.2% year-over-year, and some hedge funds have quietly entered, betting on distressed assets being reborn after interest rates decline. The poor are betting on day trading to turn their fortunes around. Zero-commission brokers combined with “signal teachers” on social media have caused retail accounts to surge by 60% compared to pre-pandemic levels.
In the perpetual contract market, funding rates have been positive for three consecutive days — longs are still paying interest to shorts, and leveraged longs refuse to back down. This scene is reminiscent of gamblers borrowing chips to double down, with only a silent "all in" in their eyes.
Silent Losses and a Decent Exit
No one likes to talk about losses, but data doesn’t lie. In 2022, 72% of retail day traders lost money, with an average loss of about $1500 per person, and fewer than 3% were able to profit long-term. Those winners posting their trades on social media are just survivors’ bias.
The advantage of anonymous trading lies here. When you lose, you can exit gracefully without facing ridicule from group members or questions from family. It’s not cowardice; it’s leaving yourself an unseen escape route. Those whales operating under pseudonyms on cryptocurrency exchanges, after liquidation, no one knows in which corner they rebuild their positions.
Judgment: AI’s Rally Will Eventually Have to Pay the Piper
I’ll put my judgment here first: the rally supported by AI is very likely to come with a reckoning. Before the internet bubble burst in 2000, Nasdaq was also hailed as the new economic paradigm, only to retreat 78% from its peak. Today’s story has a different protagonist, but the narrative formula remains unchanged.
We’ll verify tomorrow. I bet on the former — whether it’s a slap in the face or a deification, the market will have its answer. But interestingly, the market always has the power to surprise everyone, including me.
In this era driven by algorithms, leverage, and emotions, the only way to stay clear-headed may be to admit: the frenzy can be very narrow, losses can be very silent, and all we can do is leave ourselves a path to quietly exit. #消费动能转弱，9月政策仍受通胀制约 These days, it feels like the crypto market has become a cash machine for the US stock market. But actually, no need to panic, everything will come back! Many people panic when they see BTC falling and US stocks rising, but don’t rush, the core reason is just three words: liquidity extraction. US stocks have AI earnings support, institutions are reluctant to sell, so when they need cash to cover margin calls, who do they hit first? BTC, which runs 24/7 and has the best liquidity, naturally becomThe Final Flush Nobody Is Pricing In $BTC is where things get really interesting. Everyone is positioned around the same idea: → Bottom is already in → Or maybe one final dip to ~$57K That consensus is exactly what makes me cautious. BTC is already ~50% below ATH, yet we still haven't seen the kind of fear, capitulation, or “we’re going to zero” sentiment that usually accompanies a true macro bottom. If equities finally get a meaningful correction while BTC is already this weak, the downside rThese days, it feels like the crypto market has become a cash machine for the US stock market.
But actually, no need to panic, everything will come back!
Many people panic when they see BTC falling and US stocks rising, but don’t rush, the core reason is just three words: liquidity extraction. US stocks have AI earnings support, institutions are reluctant to sell, so when they need cash to cover margin calls, who do they hit first? BTC, which runs 24/7 and has the best liquidity, naturally becomes the first cash machine.
So why the cash shortage? The root cause is the yen. Global institutions have long borrowed yen at almost zero interest, converted it to dollars to buy US Treasuries, US stocks, and leveraged BTC. But then the US and Japan jointly intervened in the exchange rate, pushing the yen up, plus expectations of Japanese rate hikes, the arbitrage institutions suddenly couldn’t hold on—losing money on exchange rates, borrowing costs rising, they had no choice but to sell assets to repay debts.
This wave has basically nothing to do with BTC fundamentals; it’s purely a short-term liquidity squeeze triggered by yen arbitrage liquidation. Once this deleveraging clears, the transmission path is actually very clear: US and Japan stabilize US Treasuries → US stocks stabilize → arbitrage unwinds → rate cut expectations rise again → money flows out, and high-elasticity assets like BTC will naturally bounce the most. Simply put, just wait for the extracted liquidity to flow back.
My long position on Ethereum is already stuck, the 1885 long order, since I opened it, ETH hasn’t even touched 1890, am I really that unlucky? $ETH, damn it, just please go up for me!!!!
$OKB okb pulled back today, a normal correction, killing some leverage. I’m planning to buy another one today, already placed an order at 103, hoping to get filled Since 7.8, three reckless trades
Dxyx was the only one in the past two months; I couldn't resist chasing the high and deserved the loss.
Beat, I always thought 1.1 was the bottom, kept a cautious eye, bought at 1, and it exploded.
Next, I thought it was 0.7, still felt unsafe, so at 0.55, I steeled myself, using stop-loss as a clue, adjusted to 0.5, but it still exploded.
For Beat, actually, there’s a lesson from Lab’s previous experience: once it drops, it’s the dog trader’s massive zero-cost chips starting to unload, a wave of wealth freedom, washing hands clean, leaving no room.
I won’t participate in such second waves that have already exploded anymore; the stop-loss line is too fragile. Indeed, there’s short-term potential for one or two times gains, but it’s too easy to get liquidated.
Using those liquidated positions to chase and secure positions in promising latent coins that haven’t exploded yet is the best choice.
Lab actually has had many waves; it feels even stronger than Rave. If nothing unexpected happens, it will be the undisputed top speculative coin this year.
But out of ten like this, nine only have that one wave; after the explosion, it’s over.
I can’t participate in such trades anymore.
For a few latent ones, either I can’t buy enough or can’t buy at all, repeatedly tormenting my mindset.
Actually, even if I put all my positions out now, it might take three to four months to see an explosion.
I’ve waited nearly two months now; maybe I can buy enough next month, and perhaps it will explode the month after.
Actually,
It’s nothing; as long as I don’t act recklessly, opportunities will always come.
The most important secret is not to watch the market every day. This current job helps me avoid watching the market daily, preventing mistakes and allowing me to be passively patient.
And occasionally glancing, keeping about half an hour intermittently each day to follow the market, accumulate potential targets, and get familiar with the rhythm—actually, that’s more than enough.New LINK signal:
Santiment recorded 246 LINK transactions worth $100,000 or more within 24 hours, the highest level in about 5 months. Whale activity increased simultaneously as LINK broke the resistance zone, indicating that large capital is entering the recovery phase, not just small retail trades.
BitGo has chosen Chainlink CCIP as the exclusive cross-chain infrastructure for WBTC, replacing the previous bridge solution.
What I care about most right now: it's not just the price increase, but the price increase + a strong rise in whale activity Bitcoin remains below the downtrend line despite the simultaneous rise of stocks and gold. What has already been priced in, and what variables have yet to be reflected? While stocks and gold hit all-time highs on the same day, Bitcoin is engaged in a battle around the $63,000 support level. SPY reached a record high of $776.94, and XAU traded at $4,279.2, up 5.02% over 24 hours, showing a trend of expanding both risk appetite and inflation hedge demand simultaneously. This suggests that dollar liquidity is being distributed simultaneously to both risk assets and safe-haven assets. The key question is whether this liquidity can also flow into Bitcoin. Currently, Bitcoin is in a structurally bearish phase, maintaining a weekly downtrend alignment and a MACD death cross. However, rebound attempts have started on the 1-hour and 15-minute charts, and the $63,000 area is acting as the maximum pain price for weekend options expiry, pulling the price. The Fear & Greed Index stands at 30, indicating a fear zone. From the perspective of fund behavior,📊 $BCH Contract Liquidation Update (August 15)
According to liquidation data, the whale has completed a comprehensive long-short squeeze on BCH across short to long cycles. Shorts have been continuously wiped out in the short cycle, leaving longs no breathing room at all...
Time Total Liquidation Long Liquidation Short Liquidation
1 hour $81.26 $81.26 $0
4 hours $2,641.82 $2,641.82 $0
12 hours $31,900 $30,100 $1,779.57
24 hours $35,200 $33,000 $2,156.03
From the $BCH liquidation data, long liquidations crush shorts in the 1-hour window, with shorts completely wiped out. The long squeeze unfolds like a textbook case but with very small volume; at 4 hours, longs continue to crush shorts, which are still fully wiped out, with the long squeeze intensity exploding at a nuclear level, liquidation volume jumping from $81 to $2,641; at 12 hours, longs still dominate, with longs 16.9 times shorts. Although the long squeeze intensity slightly weakens, it remains extremely strong, with liquidation volume soaring to $31,900; at 24 hours, longs continue to dominate, with $33,000 long liquidations versus $2,156 shorts, longs 15.3 times shorts — the whale has completed a full-spectrum long squeeze on BCH from short to long cycles. Shorts are fully wiped out at 1 and 4 hours, and although they appear in mid-to-long cycles, they are still continuously harvested, with cumulative liquidations exceeding $35,200. This is a textbook-level one-sided long squeeze, with longs controlling the market throughout all cycles. But crucially, the long-to-short domination ratio shrinks from infinite at 4 hours to 15.3 times at 24 hours, indicating the long squeeze energy is continuously depleting, and longs and shorts are returning to balance. Everyone should manage positions carefully and be alert to possible sudden reversals.
⚠️ Risk Warning: All BCH cycle long liquidations continue to crush shorts, showing highly consistent direction, but the ratio narrows from infinite at 4 hours to 15.3 times at 24 hours, with long squeeze momentum weakening. Beware of directional reversal risk; 24-hour liquidations account for 94% of the daily total, showing extreme concentration. Leverage is recommended to be compressed below 3x. Do not blindly bottom-fish; strictly control positions and wait for clear directional signals.
🔥 Market Barometer | August 15
Today's three hot topics point to the same theme: the macro "stagflation" dilemma remains unresolved, but the AI sector has already entered a new phase of "heavy capital, high valuation, rapid expansion."
📉 Weakening Consumer Momentum: No hope for rate cuts, no courage for hikes
US July retail sales fell 0.6% month-over-month, the largest drop in 14 months; University of Michigan consumer sentiment preliminary index dropped to 51, well below the expected 55. Consumer anxiety about the future is turning into actual spending contraction.
But inflation stickiness still tightly restricts policy space. July core CPI rose 2.5% year-over-year, exceeding the Fed's 2% target for the sixth consecutive year. CME data shows the probability of a rate hike in September has dropped to 28.6%, but this is not a "prelude to rate cuts," rather an awkward wait due to "lack of strength to hike." BMO analysts bluntly state retail data will "support the Fed in keeping rates unchanged next month" — no move, not because it's enough, but because they dare not move.
🤖 OpenAI vs. Anthropic Valuation Race: Bubble or Revolution?
The AI valuation race is heating up. OpenAI completed a $7 billion buyback at an $852 billion valuation, but executives are leaving one after another, and the gap between revenue and burn rate is tearing market confidence.
Meanwhile, Anthropic is expected to go public in October, with some investors valuing it as high as $2 trillion. This "sky-high" valuation is supported by its enterprise-grade large model API market share reaching 32%, surpassing OpenAI's 25%. By enterprise revenue metrics, annualized revenue is expected between $100 billion and $120 billion by the end of 2026.
A five-year-old company with a $2 trillion valuation. The market is betting not on profits but on AI's complete restructuring of the enterprise market.
🏗️ SK Hynix $720 Billion Expansion: Betting on AI Computing Power "Never Sleeping"
Storage leader SK Hynix announced a $720 billion investment to build the world's largest memory factory network, targeting HBM capacity expansion. The company clearly states: memory has upgraded from a component to a core AI infrastructure.
Can returns be realized? In Q1, SK Hynix held 58% of the HBM market share, and order visibility seems stable. But the biggest risk is the mismatch between expansion cycles and demand fluctuations — its US stock has retraced about 21% from the July peak. If AI demand growth slows, the hundred-billion capacity could turn from a "moat" into a "cost black hole."
💎 Summary
Weakening consumption, persistent inflation — macro hovers on the edge of "stagflation"; AI valuations rise from $852 billion to $2 trillion, with the market pricing next-generation enterprise technology in cold hard cash; SK Hynix bets on AI computing power demand never fading with a $720 billion expansion plan. When macro weakness, high valuations, and heavy asset expansion converge — the AI sector is moving from "storytelling" to a full "cash test" phase. #消费动能转弱，9月政策仍受通胀制约
#OpenAI与Anthropic估值竞赛升温
#海力士扩产提速，资本开支能否兑现回报 The New Capital Battlefield Under the AI Frenzy: From Valuation Divergence to the Ultimate Test of Power
The rhythm of sector rotation has never stopped. Just as the memory market welcomed a brief respite, the AI sector once again stirred up market heat. Facing this phenomenon of seamless capital switching between hotspots, a reasonable logical deduction naturally arises: after storage capacity and computing power have been successively pushed higher, will the next stop for capital turn to power? After all, no matter how powerful the algorithms or how advanced the models, without a stable power supply, everything is ultimately just an illusion.
Focusing on the current market hotspots, two Pre-IPO AI concept tokens, $ANTHROPIC and $OPENAI, have recently shown extremely active performance, with their K-line trends exhibiting highly similar correlated characteristics. This technical pattern clearly reflects that the current market is mainly driven by market sentiment and capital waves, having largely detached from fundamental constraints in the short term.
However, comparing their fundamental data reveals very different underlying business logic. Anthropic achieved revenue of 11.5 billion in Q2, with a quarter-on-quarter growth rate exceeding 100%, and has successfully turned profitable; OpenAI, although with an annualized revenue as high as 40 billion and also doubling growth, is estimated to only reach profitability by 2030. This essential difference between "already having self-sustaining revenue" and "still in a high burn phase" is directly reflected in the market valuations—the valuation gap between the two is nearly double. This