Web3包青天

Bitcoin veteran holders have started not selling—does this mean the market is close to the bottom?

On-chain data shows
👉 The sell-off volume of Bitcoin’s original holders has fallen to the lowest level in 19 months

These people are usually the earliest group to have held the coins
Their characteristics are simple

👉 They don’t buy easily
👉 And they don’t sell easily
👉 When they move, it often signals a big行情 (market move)

What’s changing now is

👉 Fewer and fewer people are selling
👉 Selling pressure is easing
👉 But prices haven’t clearly rebounded yet

This creates a very subtle situation

The market looks quiet
But liquidity is actually tightening

Historically, similar conditions often mean

👉 Panic selling pressure is nearing the end
👉 The market enters a “no-buyers period”
👉 Prices may grind along at the bottom for a long time

And a question arises

If selling pressure truly has bottomed out,
then is the next phase the final stretch of silence before a rebound,
or will it continue the prolonged period of drifting downward?

Want to understand what’s happening in the market as soon as possible? Click the profile picture to follow me

Someone on Solana has started selling data to make money. What does this mean?

Kled AI in the Solana ecosystem has just received
👉 a new investment of $3 million

Total funding has now reached $14 million

This project is doing something very special:
👉 turning human data into something that can be traded

Users can upload
• file data
• medical data
• city travel data
• everyday behavioral data
Then sell it to AI companies to train models

The platform has already collected
👉 more than 12,000 sets of structured datasets

It covers many real-life scenarios

In the future, AI training won’t rely only on what models can calculate
Instead, it will start buying real human data directly

Data is moving from a byproduct
to a tradeable asset

When your everyday behavior can also be priced and sold, where exactly are the boundaries of data?

Want to understand what’s happening in the market first? Click the profile picture to follow me

AI smashed $1.3 trillion—yet the stock prices plunged together. What happened?

Samsung and SK are reportedly preparing to invest about
👉 $1.3 trillion in AI chips and data centers

But the market reaction was completely the opposite.

Korea’s two biggest companies plan to pour huge investment into AI over the next decade:
• AI chips
• Memory
• Data centers

But the moment the news broke:
👉 Samsung fell 5%
👉 SK Hynix fell 4.5%
👉 KOSPI dropped more than 3%

While the companies are aggressively ramping up for the future,
investors are worried about returns.

AI hasn’t cooled off.
It’s that the market has started to fear “money burning too fast.”

Some institutions even said directly:
👉 This is the “AI fatigue period.”

Everyone is starting to question one thing:

When can these high-ticket investments actually pay off?

If AI turns into a long-term money-burning competition,
who will be unable to hold out first?

Want to understand what’s happening in the market first? Click the avatar to follow me

一家27年前的老公司 竟然成了日本Web3最狠玩家？

SBI刚刚宣布
👉 用467亿日元收购加密交易所 Bitbank

收完这一笔之后
它直接变成日本最大的加密交易所集团之一

SBI这家公司其实很老
成立于1999年 和阿里巴巴同一年

但它这几年在做一件事：
👉 不断收交易所
👉 不断整合加密业务
👉 把日本Web3生态“打包”

现在结果是：
• 账户数约292万
• 托管资产约68亿美元
• 交易所一个接一个被收进体系里

别人是在做项目
它是在做“加密版金融帝国”

日本Web3不是新玩家的故事
而是老金融公司在重新洗牌

🌏 更有意思的是

SBI不仅做交易所
还在做：
• 跨境支付
• 稳定币
• 做市
• 托管
• 投资Ripple Circle Kraken等
几乎是把整个行业都包进来了

当一个传统金融巨头把Web3全链路都吃下去
那所谓的“去中心化”还剩多少空间

想第一时间看懂市场发生什么？点击头像关注我

一个KOL手里的币涨到7000万美元 这到底说明了什么？

加密KOL Ansem的钱包被监测到
持有的 6.04 亿枚 ANSEM 目前价值已经超过 7100 万美元

随着 ANSEM 价格上涨
他早期持有的大量代币现在被放大成一个很夸张的数字
👉 从“内容创作者持币”
👉 变成“巨额链上财富案例”

不是他突然变富
是市场把他之前的选择放大了

链上财富有时候不是赚出来的
是行情帮你“放大出来的”

如果价格继续波动
这种“纸面财富”到底能不能真正落袋

想第一时间看懂市场发生什么 ？点击头像关注我

Arthur Hayes suddenly shows interest in a new DEX. Is this about to grab Deribit's business?

Arthur Hayes mentioned on X:
👉 Hypercall (SYN) may challenge Deribit

He said that the options DEX track
has reached the stage
where new competitors must emerge.

His core judgment right now is:

👉 The Hyperliquid ecosystem is still strong,
but the real opportunities are in the "asymmetric tracks."

Options DEX is one of them.

Hypercall's positioning is straightforward:
👉 Trade on-chain options
👉 Compete with a centralized giant like Deribit

If it takes off,
it's essentially bringing the traditional options market onto the blockchain.

The point here is not the project itself,
but a signal.

The market is starting to shift its attention
from "spot trading"
to the "derivatives market."

Once options are moved on-chain and grow,
liquidity and risk structures will both change.

If Deribit is truly challenged,
then the rules of the next round of trading markets
may be different.

Want to understand what the market is doing first? Click the profile picture to follow me.

CZ suddenly clarified the three main reasons for the 2026 bear market—but the focus isn’t actually on the bear market?

In a recent interview, Binance founder CZ mentioned
👉 If 2026 enters a bear market, three main things will be driving it

He summarized three sources of pressure:
• AI is siphoning market capital
• Geopolitics makes capital more cautious
• The crypto market itself also follows a four-year cycle pattern

He also mentioned a key change:
👉 Binance.US may connect to the main site’s liquidity
Capital isn’t disappearing
It’s being rerouted from the crypto market into AI and other safer assets

Whether it’s a bear market or not isn’t the point
What matters is that the money is now relocating

If the U.S. market really does open up liquidity,
will the way crypto is priced be rewritten?

Want to understand what’s happening in the market first? Click the profile icon to follow me

AI 现在不是在起飞 反而可能是在埋雷？

央行的央行国际清算银行BIS发出警告
👉 这波 AI 万亿美元投资热潮 可能最后变成投资萧条

BIS说 大公司在疯狂砸钱做 AI 基建算力数据中心
但如果未来回报不如预期
👉 资金可能突然收紧
👉 投资热潮变成长期低迷
👉 甚至冲击全球市场

现在大家在抢着买AI未来
但没人保证真的能赚回来

AI现在看起来很热
但本质是靠预期在撑

如果回报跟不上投入
这轮AI热潮会不会突然降温

想第一时间看懂市场发生什么？ 点击头像关注我

BTC 又跌破 6 万美元 是谁在悄悄撤退？

比特币今天再一次跌破 60000 美元
日内下跌约 1.43%

📌 发生了什么
BTC 短线再次走弱
价格重新回到 6 万下方震荡
市场现在的状态就是
👉 没有强势反弹
👉 也没有明确止跌信号
👉 资金在慢慢观望
💡 人话就是
不是暴跌
但就是有人在一点一点退出风险

🧠 一句话
BTC现在不是崩
是慢慢失去冲劲

❓问题是
6 万美元到底是支撑
还是只是中途停一下的站点

想第一时间看懂市场发生什么？ 点击头像关注我

🤔 加密行业不是炒币了？怎么开始给AI和能源“当银行”了？

Framework Ventures 说了一件事：
👉 加密行业下一阶段，不再只是炒币
👉 而是变成 AI、机器人、能源的融资工具
📌 现在链上有超过 3000 亿美元稳定币
这些钱开始被用来：
• 给 AI 算力融资
• 给机器人公司供资金
• 给能源项目“提前买单”
💡 人话就是：
以前是“币涨不涨”
现在可能是“产业缺不缺钱”
🧠 一句话：
加密正在从交易市场，变成全球资金管道。

❓但问题是：
如果钱都去现实世界了，币圈自己还靠什么涨？

想第一时间看懂市场发生什么？点击头像关注我。

💰 *OpenAI IPO delayed until 2027? Saying a trillion-dollar valuation while stepping on the brakes*
According to a report by The New York Times, inside OpenAI there is discussion of:
📌 *The IPO may be pushed back to 2027*
The reason is straightforward:
👉 The market is too unstable right now to tell a “high-valuation” story.
---
💡 What’s truly interesting about this isn’t the “delay,” but the fact that *two different voices have emerged internally*:
---
📈 *One side is accelerating:*
CEO Sam Altman’s goal is very clear:
* IPO valuation target: *1 trillion*
* Current private-placement valuation: about *$852 billion*
* The aim is to lift it one more tier
His logic is:
> AI is a super cycle, so the valuation should match the future—not the present.
---
🧯 *The other side is hitting the brakes:*
The CFO is more realistic:
* Data center construction spending: *$60 billion+*
* The company is still operating at a high loss
* 2026 is still expected to be a massive deficit
Her concern is simple:
> Cash flow is still burning, and the market may not be willing to sign off on it.
---
📊 A more realistic backdrop is:
* Tech stocks are becoming more volatile
* Interest rates remain relatively high
* AI capital expenditures are entering a “cash-burning peak”
* Retail sentiment is unstable
One-sentence summary:
> It’s not that no one believes in AI anymore—it’s that “the price is hard to sell.”
---
🧠 One-sentence translation:
*OpenAI’s problem isn’t whether it’s worth a trillion—it’s that if it goes public now, will others only pay half-price.*
---
💬 More plainly:
Altman talks about the future
The CFO calculates the cash flow
Wall Street looks at the risk
The three people aren’t watching the same timeline.
---
⚠️ No predictions, no calling a direction—only breaking down the information.
The real conflict is never about “whether to list,” but about whether the market is willing to pay in advance for the future.
---
Click the avatar to follow me!

🧪 *Ethereum has updated again, but this time it feels more like an “engineer’s carnival”*
Ethereum core developers reveal:
📌 *Glamsterdam devnet-6 has been released*
The testnet continues to move forward, introducing a series of EIP updates and adjustments.
---
💡 If you see a bunch of numeric codes and feel overwhelmed, don’t panic.
In simple terms, this update boils down to three things:
---
⚙️ *1｜Make validation and build workflows more standardized (related to ePBS)*
Introduces EIP-8282 (ePBS Builder Execution Requests)
Adds a new system contract
👉 Goal:
Make block building and validation mechanisms clearer and more standardized
---
⚙️ *2｜Keep “rewriting the underlying fee/billing rules”*
Includes multiple EIPs:
* Repricing execution costs (gas-related optimizations)
* Adjusting execution-layer logic
* Optimizing how resource consumption is calculated
👉 Simply put:
> It’s not changing features—it’s “recalculating the books.”
---
⚙️ *3｜Ongoing cleanup and fine-tuning of older rules*
For example:
* Adjustments to the refund mechanism
* Execution logic optimizations
* Keeping or modifying some experimental proposals
👉 The essence is that the system becomes “increasingly more finely tuned.”
---
🧠 One-sentence translation:
> Ethereum isn’t “launching new features”—it’s continuously rewriting the rules of its own operating system.
---
📌 But here’s one real-world thing to keep in mind:
These devnet / EIP updates have a few common characteristics:
* Lots of changes, but weak user-visible impact
* Progresses over long cycles—not driven by short-term market trends
* More like slow “infrastructure tuning”
---
Click the avatar to follow me!

🤖 *AI starts making money, but is it “just enough to pay the electricity bill”?*
Research from Exponential View shows:
📊 In Q1 2026 (excluding China)
AI’s global revenue is about *$25 billion*
— for the first time, it exceeds the *$21 billion* depreciation cost for the same period.
On the surface, this looks like a milestone:
> AI has finally “earned more than it spends.”
But the issue is—things are not that simple.
---
💡 How fragile is this “surpass”?
The key assumption is that GPU / data-center equipment is depreciated over *6 years*.
But in reality:
* New-generation chips are iterating faster
* Old equipment may be retired early
* Depreciation costs can be “re-inflated” at any time
In other words:
👉 Today is “barely winning”
👉 Tomorrow it might be “mathematically invalid” again
---
💸 Even more dramatic is the other side of the账:
* Meta + Microsoft new data center commitments: on the order of *hundreds of billions of dollars*
* The industry’s overall infrastructure obligations: about *$850 billion*
And the corresponding AI quarterly revenue is:
👉 $2.5 billion
---
🧠 One-line translation:
> For now, it’s not really a question of whether AI is making money—it’s whether the money is enough to recoup the investment slowly over time.
---
📉 But the pressure doesn’t come only from the cost side:
* Cheaper models (like DeepSeek) start diverting users
* AI service prices may keep getting pushed down
* Per-user profit margins are being squeezed
So what happens is:
> Revenue is rising, but profit margins may be getting narrower.
---
🧠 A more realistic one-sentence summary:
*AI is like a super project just starting to break even—but the journey isn’t over yet, and the bills are already piling up to the ceiling.*
---
⚠️ No predictions, no calls for a direction—just information breakdown.
What really matters isn’t whether it’s making money, but whether this curve can keep supporting the next round of investment.
---
Click the avatar to follow me!

📉 STRC was originally described as “steady returns,” but now it’s increasingly starting to look like a Bitcoin shadow
Latest data shows:
📊 Strategy’s STRC preferred shares
Its 90-day correlation with Bitcoin has risen to 0.70
This is its highest level since listing.
💡 What does that mean?
In short:
STRC no longer feels like a “fixed-income product” as much as a “BTC leveraged shadow asset.”
📉 The price action is also very straightforward:
• STRC is down about 23% this month
• BTC is down about 20% over the same period
• STRC has fallen to around $76 (below face value)
• The discount is about 24%
On the surface, it still looks like it offers “11.5% dividends,”
but the price has started to reprice risk through volatility.
🧠 The key change lies here:
The product’s original design logic was:
👉 stable returns + bond-like characteristics + Bitcoin exposure
But now the market is pricing it as:
“high dividend + high volatility + BTC-synchronized risk”
⚙️ The deeper issue is Strategy’s underlying model:
• Issue STRC → raise funding
• Buy BTC → increase holdings
• Use dividends + additional issuance to sustain the cycle
But when the price falls below face value:
👉 the efficiency of additional issuance drops
👉 financing costs rise
👉 the ability of BTC buy-side demand gets weakened
forming a feedback loop.
💬 One-sentence translation:
It used to be a “yield collection product,” but now it’s increasingly like a “Bitcoin amplifier.”
📊 The market’s disagreement is also very clear right now:
🟢 Optimists:
Discount + high dividend = an opportunity
🔴 Cautious camp:
What you’re getting isn’t stable returns—it’s “BTC volatility packaged up”
⚠️ No predictions, no calls on direction—just information breakdown.
When “fixed income” starts to dance with risk assets, its name stops mattering.
Click my profile picture to follow me!

🤖 Musk admits: X’s algorithm—completely overhauled.
Recently, AI researcher Andrej Karpathy shared his experience using AI on X.
The result wasn’t a technical discussion, but rather a swarm of unfamiliar netizens surrounding him—he was even accused of “getting paid for promotion.”
He later lamented:
Now, X makes it easier to see anger, insults, and division than in the past, because this kind of content is easier to gain traction.
In response, Musk replied with a single line:
“We need to completely reform the algorithm.”
📌 What’s the real focus of this?
Not who got criticized.
But the fact that even the platform’s owner admits:
The current recommendation algorithm may be amplifying emotions rather than high-quality content.
💬 In plain language:
If anger is the easiest to get clicks,
then the algorithm will keep pushing anger.
Over time,
the world everyone sees will become even more full of emotion.
🧠 One-sentence summary:
Most of the time, it’s not the internet getting noisier—it’s the algorithm making the loudest people easier to notice.
⚠️ No predictions, no hype—just breaking down information.
Sometimes, changing a platform’s ecosystem isn’t about changing users, it’s about changing the algorithm.

Click the avatar to follow me!

🏦 Traditional Finance Starts “On-Chain Inventory”? Invesco Applies for a Stablecoin Fund
Asset management giant Invesco (managing assets of about $2.45 trillion) has filed an application with the U.S. SEC:
📌 Proposed: Invesco Stablecoin Reserves Onchain Fund
Simply put, this isn’t a regular fund—it’s a “fund pool for stablecoins.”
💡 What is the fund doing?
It mainly invests in:
• U.S. Treasury bonds
• Repurchase agreements (Repo)
• Cash equivalents
The goal is straightforward:
👉 Keep $1 stable
👉 And still earn interest
But the key is not just “investing”—it’s that:
🧠 Fund shares will be put on-chain (tokenized)
In other words, traditional funds + blockchain systems are starting to merge.
📌 Why is this important?
If I explain it in plain language:
Stablecoins aren’t “crypto tools” anymore—they’re becoming “Wall Street infrastructure.”
Previously, stablecoins were for trading.
Now they’re turning into:
👉 a fund custody layer
👉 an yield-generating layer
👉 a compliant reserves layer
💬 More bluntly:
It’s not the crypto world moving closer to traditional finance.
It’s traditional finance “redesigning the underlying structure” of stablecoins.
🧠 One-sentence summary:
Stablecoins are upgrading from a “medium of exchange” into a “part of the financial system.”
⚠️ No predictions, no calls for direction—just information breakdown.
The real change has never been price volatility; it’s been the fund structure that starts to change.

Click my avatar to follow me

#Invesco

🐋 A “Giant Whale-scale Multi-Head,” being repeatedly taught by the market
On-chain monitoring shows that:
An address known as the “Hyperliquid Largest Bull” is continuously adding to its position.
📊 Current holdings:
About 120,000 ETH long contracts
About 2,500 BTC long contracts
Total position size around $445 million
📉 Current unrealized loss is about: $110 million
More importantly, he’s still adding:
👉 When BTC drops to around 59,000
👉 He still uses multiple wallets to open more longs at $59,261
💡 What is he doing?
Simple translation:
Not “guessing the bottom,” but “adding more to his conviction, again and again.”
The average entry price for his BTC longs is about $69,560
The average entry price for his ETH longs is about $2,261
Meaning the market price now is:
👉 Clearly below his cost range
🧠 The key isn’t how much he’s losing, but the behavior itself:
• It indicates there are strong-conviction bulls in the market
• It suggests someone is “buying the dip harder as it falls”
• And it also shows the current volatility isn’t a one-way consensus market
But at the same time, it exposes a harsh reality:
In a high-leverage market, “conviction” doesn’t equal a margin of safety.
📉 The other side is the market’s current condition:
• Price keeps pushing lower
• Long positions get liquidated frequently
• Unrealized losses keep expanding
• Sentiment is in extreme tug-of-war
💬 One-sentence summary:
Someone is sticking to their direction, while someone else is paying the price for that persistence.
⚠️ No predictions, no calls—just information breakdown.
Markets never lack opinions; what they lack is whether those opinions can survive until the end.
Tap the avatar to follow me!

🚨 Crypto world is having another “battle royale”?
This morning, when many people opened the market chart, their first reaction was:
“ I slept for a bit—what on earth happened?”
📉 BTC briefly fell to around $58,800
📉 ETH broke below $1,565
📉 In the past 24 hours, liquidations across the whole network totaled $887 million, and more than 80% were long positions.
Don’t be scared just by “$800 million in liquidations.”
The real issue isn’t how much the price dropped, but that too many people used leverage.
When the market dips, leveraged positions are like dominoes—one after another gets liquidated—so the liquidation amount keeps growing.
This drop was mainly driven by several factors:
📌 Ongoing outflows of funds from Bitcoin ETFs
📌 Market expectations that high interest rates will last longer
📌 A pullback in U.S. tech stocks, putting pressure on risk assets across the board
📌 Excessive leverage triggering a chain of liquidations
💬 To sum it up in plain human language:
Market declines aren’t scary by themselves—the scary part is that many people think, “This time it definitely won’t drop again,” and then set leverage higher than their confidence.
⚠️ I’m not making predictions or calling direction—just breaking down the information.
Markets move every day, but understanding what’s happening matters more than guessing the next candlestick.
Click my avatar to follow me!

🧾 Web3 Bao Qingtian｜Why can’t the new GPT model be released all at once?
According to media reports, OpenAI’s next-generation model GPT-5.6 may not be released fully and directly. Instead, it will likely be tested first with a small number of partner organizations, and then gradually opened up to more users.
Many people’s first reaction is:
“Why is AI starting to do whitelists too?”
In fact, the reasons aren’t that complicated.
In the past, what people competed on was:
👉 Who had the smarter model.
Now, what people compete on is:
👉 Whose model is more stable, safer, and less likely to cause trouble.
Put simply, AI is no longer just about building chatbots—more and more companies and even governments are using it.
Once a model answers incorrectly or malfunctions, the impact could be even bigger than a server going down.
🧠 Bao Qingtian’s one-sentence translation:
In the past when releasing AI:
“Everyone, come try it!”
Now when releasing AI:
“Test it with a few people first—don’t roll it out and have it crash right away.”
This also shows that AI competition has entered a new phase.
In the future, it won’t only be about how big the parameters are, but about who is more reliable and who is more trustworthy.
⚠️ No exaggeration, no hype—just information breakdown.
Tap the avatar to follow me!

Why is Bitcoin down again?

It has gone to a place that’s “better at telling stories.”
📉 What happened?
Bitcoin has dropped by about 30% in the first half of this year,
sliding all the way from its highs to below $60,000 at one point.
But at the same time:
AI and chip stocks are being chased by capital
and the US stock market is still “pretending to correct, while actually making money.”
One sentence summary:
The money didn’t disappear—it just switched lanes.
💵 How strong is the US dollar, and what does it have to do with this?
Here’s the situation right now:
The Fed is leaning hawkish (not in a hurry to cut rates). High interest rates mean money prefers to stay in banks to earn interest.
That’s awkward for assets like Bitcoin—“non–interest-bearing” ones.
In plain language:
People used to think “printing money = buy Bitcoin.”
Now it’s more like: “money is expensive—don’t be impulsive.”
📊 Why are institutions backing out too?
Several consecutive weeks of outflows from ETFs show one thing:
Institutions aren’t quitting—they’re lining up for AI first.
Capital flows are very practical:
AI: has a growth story
Bitcoin: doesn’t have a fresh script yet
🧠 One-line translation of Bao Qingtian:
The market isn’t punishing Bitcoin—it’s rewarding the people who can “tell the future better.”
⚠️ No predictions, no calling directions—just breaking down information.
The market never really cares who’s right.
It only cares who the money wants to hear tell a story today.
Tap the profile picture to follow me!