大晖哥
X@Whdysseus ｜1783DAO & 博链BroadChain 创始人｜2025币安全球区块链百强创作者(入围) ｜全球宏观与策略分析师(野生)｜前证券与基金从业｜Crypto OG since 2014｜关注Crypto、美股与港股市场 DYOR
34 Following
1.8K+ Followers
2.0K+ Liked
74 Shared
Posts
All
Quotes
PINNED
Faced with Binance's sweep and bloodbath, compliant (onshore) trading platforms can only watch helplessly.
As for other offshore platforms, there aren't many effective response plans available.
Chinese people emphasize 'timeliness and trends'; the only thing offshore platforms can do is meticulous operation, honing their internal skills, and looking for those niche opportunities that Binance does not want to pursue, is unwilling to pursue, or has temporarily overlooked.
Fortunately, the exchange industry is somewhat like securities companies— as long as users continue to trade, they can survive long-term and continue to make profits.
Having dozens or even hundreds of platforms coexisting is the norm.
But now, the scale of Binance and #BNB is already ten times larger than that of all other offshore platforms combined.
The industry concentration has reached a historical record, known as 'one super and many strong'.
In fact, Binance's journey to today is certainly not due to luck; once, around 2023, at the Binance Chinese meeting in Dubai, He Yi mentioned the transition from 'good to great'. I also discussed this issue with Yi Jie, asking how to achieve the transition from 'good to great'; currently, Binance has achieved this 100%.
Since its founding, Binance has made the right choices at almost every key juncture and executed them nearly perfectly:
Founder CZ is a Canadian-Chinese engineer, and co-founder Yi Jie is a marketing expert deeply connected with CZ;
At the end of 2017, decisively went overseas, rising to industry leader from spot trading;
Acquired the old team of OKX, quickly filling the contract gap;
Launched the BNB public chain, turning the platform token into a public chain ecosystem;
Headquarters relocated to Dubai, securing investment and protection from the Abu Dhabi Sovereign Fund;
Attracting traditional financial institutions and listed companies to include BNB in their treasury, further expanding the BNB ecosystem.
Now, Binance and BNB only need to be accountable to Middle Eastern investors (Dubai, Abu Dhabi);
They neither need to heed the U.S. nor the Chinese government's opinions—of course, excluding counter-terrorism and anti-money laundering regulations.
Other offshore platforms, however, are still being pursued and blocked by regulators from various countries.
The 'destructive power' of Crypto is too great; aside from the U.S. and China, even Singapore and Hong Kong find it difficult to fully digest.
As for other offshore platforms, there aren't many effective response plans available.
Chinese people emphasize 'timeliness and trends'; the only thing offshore platforms can do is meticulous operation, honing their internal skills, and looking for those niche opportunities that Binance does not want to pursue, is unwilling to pursue, or has temporarily overlooked.
Fortunately, the exchange industry is somewhat like securities companies— as long as users continue to trade, they can survive long-term and continue to make profits.
Having dozens or even hundreds of platforms coexisting is the norm.
But now, the scale of Binance and #BNB is already ten times larger than that of all other offshore platforms combined.
The industry concentration has reached a historical record, known as 'one super and many strong'.
In fact, Binance's journey to today is certainly not due to luck; once, around 2023, at the Binance Chinese meeting in Dubai, He Yi mentioned the transition from 'good to great'. I also discussed this issue with Yi Jie, asking how to achieve the transition from 'good to great'; currently, Binance has achieved this 100%.
Since its founding, Binance has made the right choices at almost every key juncture and executed them nearly perfectly:
Founder CZ is a Canadian-Chinese engineer, and co-founder Yi Jie is a marketing expert deeply connected with CZ;
At the end of 2017, decisively went overseas, rising to industry leader from spot trading;
Acquired the old team of OKX, quickly filling the contract gap;
Launched the BNB public chain, turning the platform token into a public chain ecosystem;
Headquarters relocated to Dubai, securing investment and protection from the Abu Dhabi Sovereign Fund;
Attracting traditional financial institutions and listed companies to include BNB in their treasury, further expanding the BNB ecosystem.
Now, Binance and BNB only need to be accountable to Middle Eastern investors (Dubai, Abu Dhabi);
They neither need to heed the U.S. nor the Chinese government's opinions—of course, excluding counter-terrorism and anti-money laundering regulations.
Other offshore platforms, however, are still being pursued and blocked by regulators from various countries.
The 'destructive power' of Crypto is too great; aside from the U.S. and China, even Singapore and Hong Kong find it difficult to fully digest.
PINNED
Binance VIP 9 Customer Exclusive
Global Limited Edition of 50 Sets
Texas + Mahjong Set
The family of 1783DAO is so powerful
Global Limited Edition of 50 Sets
Texas + Mahjong Set
The family of 1783DAO is so powerful
I think many people overlook a problem:
Social collaboration and organizational structure is an extremely complex system.
Now, the 'One Person Company' (OPC)
is not much different from the 'mass entrepreneurship and innovation' around 2015.
Starting a business without considering the business model will ultimately lead to a mess.
The greater probability in the future is not that 'One Person Companies' will bloom everywhere, but rather that more people will be unemployed.
AI does not replace jobs; it replaces nodes in the chain of socialized collaboration. When a large company starts using AI to cut out the middle layers, those who are released will not automatically become the bosses of 'One Person Companies', but rather become people without income.
There will definitely be a small group of people who can successfully run the 'One Person Company' model. For example:
- Tech wizards who create products alone and live on subscription income or sell to large companies, such as OpenClaw.
- Super individuals who take orders based on personal branding, with extremely high profit margins, such as self-media and influential figures.
But this is just like 'mass entrepreneurship': what you see are the survivors; no one mentions those who died within three months.
In the context of the significant productivity improvement brought by AI large models and high-performance robots, the tools have become stronger, but the speed of iteration has become faster, and the life cycle shorter.
The biggest pitfall of 'One Person Companies' is: it leads us to believe that we can succeed without dealing with complex systems. But the real world has never operated this way.
Social collaboration and organizational structure is an extremely complex system.
Now, the 'One Person Company' (OPC)
is not much different from the 'mass entrepreneurship and innovation' around 2015.
Starting a business without considering the business model will ultimately lead to a mess.
The greater probability in the future is not that 'One Person Companies' will bloom everywhere, but rather that more people will be unemployed.
AI does not replace jobs; it replaces nodes in the chain of socialized collaboration. When a large company starts using AI to cut out the middle layers, those who are released will not automatically become the bosses of 'One Person Companies', but rather become people without income.
There will definitely be a small group of people who can successfully run the 'One Person Company' model. For example:
- Tech wizards who create products alone and live on subscription income or sell to large companies, such as OpenClaw.
- Super individuals who take orders based on personal branding, with extremely high profit margins, such as self-media and influential figures.
But this is just like 'mass entrepreneurship': what you see are the survivors; no one mentions those who died within three months.
In the context of the significant productivity improvement brought by AI large models and high-performance robots, the tools have become stronger, but the speed of iteration has become faster, and the life cycle shorter.
The biggest pitfall of 'One Person Companies' is: it leads us to believe that we can succeed without dealing with complex systems. But the real world has never operated this way.
In the past seven years, the changes have been significant.
In 2017, discussions were about whether it was possible to buy Crypto.
By 2026, discussions shifted to whether it was possible to use Crypto to buy a house.
Crypto assets are transitioning from 'speculative products' to 'collateralizable assets'.
This time, it's related to the U.S. real estate sector’s state-owned enterprise—Fannie Mae, which was the main character in the 2008 'subprime mortgage crisis', surprisingly wants to engage in Crypto-backed mortgages.
Mortgage service provider Better Home (NASDAQ:BETR) and Crypto exchange Coinbase (NASDAQ:COIN) have collaborated to launch a new product:
When you apply for a mortgage to buy a house, you can use your Crypto assets as collateral for the down payment without having to sell them for cash.
Although using Crypto assets to buy a house is not a novel idea and similar products have been available in the market for a while, the key this time is Fannie Mae's involvement.
The ideal scenario for this play is:
Buying a house → Using Crypto assets as collateral for the down payment → Obtaining a mortgage → The house appreciates + Crypto assets still in hand.
Fannie Mae, to put it bluntly, is the 'national team' of the U.S. mortgage market.
It does not lend money directly to homebuyers; instead, it buys mortgages from banks, allowing banks to have funds to continue lending.
It's essentially a 'funding pump' for the mortgage market, ensuring that homebuyers can borrow money when purchasing a house.
Founded in 1938, it has the backing of the U.S. federal government, is regulated by the Federal Housing Finance Agency, and primarily serves low to moderate-income families.
The 2008 'subprime mortgage crisis' plunged many U.S. financial giants into trouble. Fannie Mae and Freddie Mac were taken over by the government, and Lehman Brothers went bankrupt, becoming a hallmark of the global financial crisis.
This time, Fannie Mae's entry to support Crypto-backed mortgages sends a clear signal: the mainstream financial system is starting to take Crypto assets seriously.
Xiao Huihui believes this is a tangible benefit for the Crypto industry, but is also concerned that Fannie Mae might mismanage risks and mess up, after all, they had a slip-up ten years ago.
In 2017, discussions were about whether it was possible to buy Crypto.
By 2026, discussions shifted to whether it was possible to use Crypto to buy a house.
Crypto assets are transitioning from 'speculative products' to 'collateralizable assets'.
This time, it's related to the U.S. real estate sector’s state-owned enterprise—Fannie Mae, which was the main character in the 2008 'subprime mortgage crisis', surprisingly wants to engage in Crypto-backed mortgages.
Mortgage service provider Better Home (NASDAQ:BETR) and Crypto exchange Coinbase (NASDAQ:COIN) have collaborated to launch a new product:
When you apply for a mortgage to buy a house, you can use your Crypto assets as collateral for the down payment without having to sell them for cash.
Although using Crypto assets to buy a house is not a novel idea and similar products have been available in the market for a while, the key this time is Fannie Mae's involvement.
The ideal scenario for this play is:
Buying a house → Using Crypto assets as collateral for the down payment → Obtaining a mortgage → The house appreciates + Crypto assets still in hand.
Fannie Mae, to put it bluntly, is the 'national team' of the U.S. mortgage market.
It does not lend money directly to homebuyers; instead, it buys mortgages from banks, allowing banks to have funds to continue lending.
It's essentially a 'funding pump' for the mortgage market, ensuring that homebuyers can borrow money when purchasing a house.
Founded in 1938, it has the backing of the U.S. federal government, is regulated by the Federal Housing Finance Agency, and primarily serves low to moderate-income families.
The 2008 'subprime mortgage crisis' plunged many U.S. financial giants into trouble. Fannie Mae and Freddie Mac were taken over by the government, and Lehman Brothers went bankrupt, becoming a hallmark of the global financial crisis.
This time, Fannie Mae's entry to support Crypto-backed mortgages sends a clear signal: the mainstream financial system is starting to take Crypto assets seriously.
Xiao Huihui believes this is a tangible benefit for the Crypto industry, but is also concerned that Fannie Mae might mismanage risks and mess up, after all, they had a slip-up ten years ago.
Are storage chips facing a short-term valuation kill, or is the logic really going to be reassessed?
SanDisk (NASDAQ: SNDK), Micron Technology (NASDAQ: MU), Western Digital (NASDAQ: WDC), Seagate Technology (NASDAQ: STX) and other US stock storage chip sectors have started to continuously pull back.
The most innovative giant company in the universe, Google, is causing a stir again.
This time it's not an AI model, it's a chip storage module.
Google just introduced a compression algorithm called TurboQuant, claiming it can save about 6 times the memory.
It's equivalent to being able to fit six copies in the space that originally held one data set.
If the compression algorithm can boost storage efficiency to this extent, the ceiling for hardware demand may need to be recalculated.
Previously it was: AI computing power increases → storage demand grows linearly → expand, expand, expand.
Now it is: the algorithm helps you save a large portion first, and the logic of hardware demand has changed.
It's not that storage chips are no longer needed, but that the "expectation gap" has been shattered.
SanDisk (NASDAQ: SNDK), Micron Technology (NASDAQ: MU), Western Digital (NASDAQ: WDC), Seagate Technology (NASDAQ: STX) and other US stock storage chip sectors have started to continuously pull back.
The most innovative giant company in the universe, Google, is causing a stir again.
This time it's not an AI model, it's a chip storage module.
Google just introduced a compression algorithm called TurboQuant, claiming it can save about 6 times the memory.
It's equivalent to being able to fit six copies in the space that originally held one data set.
If the compression algorithm can boost storage efficiency to this extent, the ceiling for hardware demand may need to be recalculated.
Previously it was: AI computing power increases → storage demand grows linearly → expand, expand, expand.
Now it is: the algorithm helps you save a large portion first, and the logic of hardware demand has changed.
It's not that storage chips are no longer needed, but that the "expectation gap" has been shattered.
OpenAI's new round of fundraising exceeds $120 billion.
a16z, DE Shaw, Abu Dhabi MGX, TPG, and Rowe Price (NASDAQ:TROW) participated in the investment, and Microsoft also participated in the follow-on investment.
OpenAI predicts that by the end of 2026, the proportion of To B and To C business will shift to a fifty-fifty split.
Additionally, the video generation application Sora, which was launched with high profile at the end of September 2025 and briefly topped the App Store charts, announced its shutdown six months later.
OpenAI is stabilizing its revenue structure with To B while experimenting, eliminating, and consolidating on the To C side.
a16z, DE Shaw, Abu Dhabi MGX, TPG, and Rowe Price (NASDAQ:TROW) participated in the investment, and Microsoft also participated in the follow-on investment.
OpenAI predicts that by the end of 2026, the proportion of To B and To C business will shift to a fifty-fifty split.
Additionally, the video generation application Sora, which was launched with high profile at the end of September 2025 and briefly topped the App Store charts, announced its shutdown six months later.
OpenAI is stabilizing its revenue structure with To B while experimenting, eliminating, and consolidating on the To C side.
Polymarket made two major adjustments, providing a lot of information:
First: The range of fees charged has expanded, mainly aimed at increasing liquidity.
Starting from March 30, the taker fee was previously only applicable to the Crypto and Sports categories, but now Finance, Politics, Economics, Culture, and Weather have been added.
The new rates are not fixed; they are dynamically calculated, with peak values generally around 50% probability (meaning the most competitive positions). The maximum for the Crypto category is about 1.8%, while the other newly added categories are around 1%.
The collected fees are mainly used to incentivize market makers to place orders. Fees are calculated in USDC, deducted from shares when buying and from USDC when selling.
Second, user growth, the referral program (commission plan) has also been opened up.
Previously in beta testing, now all users with trading volumes exceeding $10,000 can participate.
The more new users trade, the more rewards they receive, calculated proportionally.
First: The range of fees charged has expanded, mainly aimed at increasing liquidity.
Starting from March 30, the taker fee was previously only applicable to the Crypto and Sports categories, but now Finance, Politics, Economics, Culture, and Weather have been added.
The new rates are not fixed; they are dynamically calculated, with peak values generally around 50% probability (meaning the most competitive positions). The maximum for the Crypto category is about 1.8%, while the other newly added categories are around 1%.
The collected fees are mainly used to incentivize market makers to place orders. Fees are calculated in USDC, deducted from shares when buying and from USDC when selling.
Second, user growth, the referral program (commission plan) has also been opened up.
Previously in beta testing, now all users with trading volumes exceeding $10,000 can participate.
The more new users trade, the more rewards they receive, calculated proportionally.
Another Dimension Gala
The girl with more titles than 'Mother of Dragons'
Entering the Crypto industry, first made a high-profile entry into OKX, serving as the BD for the wallet business line
Later, she left OKX in a high-profile manner to join Binance as the BD for the BNB ecosystem, and soon left Binance, with rumors suggesting she was fired; however, Another Dimension Gala had He Yi testify on X, and He Yi denied that she was fired.
But after being suspected of being subjected to continuous 'online violence', she withdrew for a long time, and now, in the context of the ongoing bear market, she has joined OKX again, taking on a strategic role in the president's office.
It seems that OKX needs someone who understands the US market and speaks Mandarin, and Another Dimension appears to be a good candidate.
The girl with more titles than 'Mother of Dragons'
Entering the Crypto industry, first made a high-profile entry into OKX, serving as the BD for the wallet business line
Later, she left OKX in a high-profile manner to join Binance as the BD for the BNB ecosystem, and soon left Binance, with rumors suggesting she was fired; however, Another Dimension Gala had He Yi testify on X, and He Yi denied that she was fired.
But after being suspected of being subjected to continuous 'online violence', she withdrew for a long time, and now, in the context of the ongoing bear market, she has joined OKX again, taking on a strategic role in the president's office.
It seems that OKX needs someone who understands the US market and speaks Mandarin, and Another Dimension appears to be a good candidate.
The market is pricing in "interest rate hike risks," but it is still far from the stage where "the interest rate hike cycle has begun." All risk assets are retreating; the core issue is not sentiment, but rather the changing macroeconomic environment. Oil prices remain high in the long term, and inflation expectations are rising again, which is the most damaging logic in the current market. More importantly, the market is starting to re-evaluate the possibility of "interest rate hikes" rather than rate cuts. Current interest rate expectations: April rate hike probability of 25bp: 12.4% June cumulative rate hike probability of 25bp: 21.9% Maintaining the interest rate unchanged remains mainstream (approximately 76.5%) This indicates one thing: rate hikes are not the main storyline, but have shifted from "0 expectations" to "risks that need to be priced in." Once the market starts pricing in rate hikes, it will directly suppress valuations (especially for Nasdaq, tech stocks, AI concept stocks + Crypto). Another overlooked point: the new Federal Reserve Chair will not take office until after June, and even with a change, it does not mean an immediate shift to rate cuts. Monetary policy has never been determined by one person, but rather by the consensus under the long-term systematic operation of the entire Federal Reserve.
History is repeating itself
Internet → Replacing newspapers
Mobile Internet → Replacing websites
AI → Restructuring the entry, apps are gradually disappearing
The timeline is becoming clearer:
2022's OpenAI, ChatGPT = iPhone moment, capability breakthrough
In 2026, OpenAI, ChatGPT plus OpenClaw = more like iOS, becoming a super entry
Look, OpenAI plans to merge ChatGPT, Codex, and its browser into a unified desktop 'super application'.
In the past, ChatGPT was responsible for conversation chatting, Codex wrote code, and information was searched through the browser
In the future, OpenAI will be a single entry point, completing all tasks.
OpenAI is building the operating system (AI OS) for the AI era.
Internet → Replacing newspapers
Mobile Internet → Replacing websites
AI → Restructuring the entry, apps are gradually disappearing
The timeline is becoming clearer:
2022's OpenAI, ChatGPT = iPhone moment, capability breakthrough
In 2026, OpenAI, ChatGPT plus OpenClaw = more like iOS, becoming a super entry
Look, OpenAI plans to merge ChatGPT, Codex, and its browser into a unified desktop 'super application'.
In the past, ChatGPT was responsible for conversation chatting, Codex wrote code, and information was searched through the browser
In the future, OpenAI will be a single entry point, completing all tasks.
OpenAI is building the operating system (AI OS) for the AI era.
From "glorious IPO" to "massive layoffs to survive," the cryptocurrency exchange Gemini (NASDAQ:GEMI) took less than half a year, with its stock price dropping nearly 90% from its peak.
As of now, Gemini has laid off about 30% of its workforce, leaving only 445 employees, and has gradually exited the UK, EU, and Australian markets.
Cumulative losses reached as high as $585 million in 2025, with fourth-quarter revenue around $60 million, but losses continue to widen.
The core logic of the exchange is that revenue comes from trading volume, and with the current market downturn, trading volume has declined, the compliance costs for listed exchanges are high, and fixed expenses cannot be reduced, leading to continued losses.
It's not just Gemini that is retreating; Crypto.com has laid off employees, and Kraken has paused its IPO, indicating that the exchange industry is entering a contraction and deleveraging phase.
As of now, Gemini has laid off about 30% of its workforce, leaving only 445 employees, and has gradually exited the UK, EU, and Australian markets.
Cumulative losses reached as high as $585 million in 2025, with fourth-quarter revenue around $60 million, but losses continue to widen.
The core logic of the exchange is that revenue comes from trading volume, and with the current market downturn, trading volume has declined, the compliance costs for listed exchanges are high, and fixed expenses cannot be reduced, leading to continued losses.
It's not just Gemini that is retreating; Crypto.com has laid off employees, and Kraken has paused its IPO, indicating that the exchange industry is entering a contraction and deleveraging phase.
The downtrend cycle of the Crypto industry bear market and the paradigm shift of AI technology
A wave of layoffs is sweeping through the Crypto industry
Earlier in 2026, the cryptocurrency exchange Gemini (NASDAQ:GEMI) laid off 25% of its staff and closed its operations in the UK, Europe, and Australia.
Subsequently, the payment company Block (Square) (NYSE:XYZ), founded by X (Twitter), laid off nearly half of its employees.
The latest instance of layoffs is the cryptocurrency exchange Crypto.com, which laid off 12%, about 180 people, and is transitioning to AI.
The biggest logic behind this is—more can be done with fewer people.
A wave of layoffs is sweeping through the Crypto industry
Earlier in 2026, the cryptocurrency exchange Gemini (NASDAQ:GEMI) laid off 25% of its staff and closed its operations in the UK, Europe, and Australia.
Subsequently, the payment company Block (Square) (NYSE:XYZ), founded by X (Twitter), laid off nearly half of its employees.
The latest instance of layoffs is the cryptocurrency exchange Crypto.com, which laid off 12%, about 180 people, and is transitioning to AI.
The biggest logic behind this is—more can be done with fewer people.
Do you know why commodities are called - "widow maker"?
Husbands (men) all go bankrupt, lose everything, and die.
Husbands (men) all go bankrupt, lose everything, and die.
No pain, no gain.
Starting over again.
If I had known this, why did I do it in the first place?
After significant layoffs and store closures at Xibei,
Xibei founder Jia Guolong was forced to leave Xibei and began a new restaurant venture—"Tianbian Claypot Noodles" has quietly opened in the Beijing 798 Art District.
The first store opened on February 13, with an average consumption of 40 to 50 yuan, specializing in the claypot noodles from Jia Guolong's hometown in Inner Mongolia's Bayannur.
Starting over again.
If I had known this, why did I do it in the first place?
After significant layoffs and store closures at Xibei,
Xibei founder Jia Guolong was forced to leave Xibei and began a new restaurant venture—"Tianbian Claypot Noodles" has quietly opened in the Beijing 798 Art District.
The first store opened on February 13, with an average consumption of 40 to 50 yuan, specializing in the claypot noodles from Jia Guolong's hometown in Inner Mongolia's Bayannur.
Satoshi Nakamoto invented Bitcoin, but Jensen Huang is the one who turned 'computing power into an era asset'.
A Token connects two waves of technological tides.
2017, the first wave of the Crypto bull market
Miners scrambled for GPUs, graphics cards sold out overnight, and Nvidia earned an extra $300 million in one quarter thanks to Crypto.
2021, DeFi exploded
Miners began industrial-scale purchasing, and graphics card prices tripled; the Crypto business became Nvidia's second-largest source of income.
2022, Ethereum's The Merge upgrade, PoW transitioned to PoS
The demand for GPU mining instantly dropped to zero; this line ended, but the story did not.
2025, AI exploded, and AI companies began frantically acquiring: electricity, data centers, computing power.
Bitcoin mining companies had already built: data centers, power systems, cooling facilities; rebuilding these would take 3-4 years.
Thus, a historical turning point emerged: AI companies began to 'take over mining sites', signing nearly $100 billion in orders within a year.
The real significance of this event is: Crypto and AI have never been two separate lines, but two uses of the same infrastructure.
The first time: mining BTC; the second time: running AI.
Satoshi Nakamoto started it, and Jensen Huang took over. Computing power has become the hard currency of this era.
A Token connects two waves of technological tides.
2017, the first wave of the Crypto bull market
Miners scrambled for GPUs, graphics cards sold out overnight, and Nvidia earned an extra $300 million in one quarter thanks to Crypto.
2021, DeFi exploded
Miners began industrial-scale purchasing, and graphics card prices tripled; the Crypto business became Nvidia's second-largest source of income.
2022, Ethereum's The Merge upgrade, PoW transitioned to PoS
The demand for GPU mining instantly dropped to zero; this line ended, but the story did not.
2025, AI exploded, and AI companies began frantically acquiring: electricity, data centers, computing power.
Bitcoin mining companies had already built: data centers, power systems, cooling facilities; rebuilding these would take 3-4 years.
Thus, a historical turning point emerged: AI companies began to 'take over mining sites', signing nearly $100 billion in orders within a year.
The real significance of this event is: Crypto and AI have never been two separate lines, but two uses of the same infrastructure.
The first time: mining BTC; the second time: running AI.
Satoshi Nakamoto started it, and Jensen Huang took over. Computing power has become the hard currency of this era.
Very, very bad news for the Crypto industry
The second largest cryptocurrency exchange in the United States, Kraken, has announced the suspension of its IPO plan worth billions of dollars, with a valuation of up to $20 billion.
Previously, Kraken had submitted an S-1 filing to the US SEC, indicating that it was originally prepared to go public, but now it has publicly stated that it has chosen to suspend it.
Currently, the visible situation in Crypto — poor macro liquidity, capital shifting towards the AI industry chain, and declining trading volumes; for exchanges, trading volume equals revenue.
The second largest cryptocurrency exchange in the United States, Kraken, has announced the suspension of its IPO plan worth billions of dollars, with a valuation of up to $20 billion.
Previously, Kraken had submitted an S-1 filing to the US SEC, indicating that it was originally prepared to go public, but now it has publicly stated that it has chosen to suspend it.
Currently, the visible situation in Crypto — poor macro liquidity, capital shifting towards the AI industry chain, and declining trading volumes; for exchanges, trading volume equals revenue.
Oklo: The end of computing power is energy, and the end of energy is nuclear energy
If the bottleneck of AI is computing power
Then the bet made by Oklo, heavily led by OpenAI founder Sam Altman, is that the end of computing power is energy, and the end of energy is nuclear energy
then Oklo in 2025 officially shifts from 'designing reactors' to 'deploying nuclear power assets + building a closed-loop nuclear industry chain'.
The core actions are very clear:
·Initiate the construction of the first reactor in Aurora, Idaho (from 0 to 1)
·Signed a prepayment agreement for a 1.2GW clean energy park with Meta (locking in long-term demand)
·Acquire Atomic Alchemy (to strengthen isotope capabilities)
·Simultaneously promote the infrastructure of fuel, recycling, and isotopes on multiple fronts
Then the bet made by Oklo, heavily led by OpenAI founder Sam Altman, is that the end of computing power is energy, and the end of energy is nuclear energy
then Oklo in 2025 officially shifts from 'designing reactors' to 'deploying nuclear power assets + building a closed-loop nuclear industry chain'.
The core actions are very clear:
·Initiate the construction of the first reactor in Aurora, Idaho (from 0 to 1)
·Signed a prepayment agreement for a 1.2GW clean energy park with Meta (locking in long-term demand)
·Acquire Atomic Alchemy (to strengthen isotope capabilities)
·Simultaneously promote the infrastructure of fuel, recycling, and isotopes on multiple fronts
Binance has directly lowered the VIP threshold to the ankle
The BNB holding requirement for VIP 1 has been reduced from 25 BNB to 5 BNB, a decrease of 80%, approximately 3300 US dollars
The asset threshold for VIP 1 has been lowered from 500,000 USDT to a minimum of 100,000 USDT, a decrease of 80%, with BNB holdings included in total assets
The contract trading volume requirement for VIP 1 has been reduced from 15 million US dollars to 5 million US dollars, a decrease of 66.7%
Next door, OKX VIP 1 is 100,000 USDT, not sure if they will also reduce it.
The BNB holding requirement for VIP 1 has been reduced from 25 BNB to 5 BNB, a decrease of 80%, approximately 3300 US dollars
The asset threshold for VIP 1 has been lowered from 500,000 USDT to a minimum of 100,000 USDT, a decrease of 80%, with BNB holdings included in total assets
The contract trading volume requirement for VIP 1 has been reduced from 15 million US dollars to 5 million US dollars, a decrease of 66.7%
Next door, OKX VIP 1 is 100,000 USDT, not sure if they will also reduce it.
Alibaba Cloud has taken the lead in raising prices, and the increase is not small
AI computing power & storage prices have been raised by up to 34%
The superficial reason is: AI demand explosion + supply chain cost increase
But the real core is summed up in one sentence: Token call volume has exploded
What is even more worth paying attention to is the structural changes behind it
1. AI computing power has begun to enter the "supply-demand imbalance" stage
·MaaS platform demand has surged
·Inference demand (not training) has become the main force
This indicates that AI has moved from the "experimental stage" to the "mass usage stage"
2. Cloud service providers have started to do "computing power allocation"
Alibaba Cloud is already doing: prioritizing Token/inference business
This means: computing power is becoming a kind of "resource allocation"
3. The business model is changing
Previously, cloud service providers competed on: price reduction + scale
Now it has changed to: price increase + priority scheduling + profit optimization
4. Who will benefit?
·GPU/chip manufacturers (demand continues to explode)
·Cloud service providers (profit improvement)
·Leading AI applications (can access resources)
Who will be squeezed? Small and medium-sized AI startups (cannot access computing power or costs rise)
AI computing power & storage prices have been raised by up to 34%
The superficial reason is: AI demand explosion + supply chain cost increase
But the real core is summed up in one sentence: Token call volume has exploded
What is even more worth paying attention to is the structural changes behind it
1. AI computing power has begun to enter the "supply-demand imbalance" stage
·MaaS platform demand has surged
·Inference demand (not training) has become the main force
This indicates that AI has moved from the "experimental stage" to the "mass usage stage"
2. Cloud service providers have started to do "computing power allocation"
Alibaba Cloud is already doing: prioritizing Token/inference business
This means: computing power is becoming a kind of "resource allocation"
3. The business model is changing
Previously, cloud service providers competed on: price reduction + scale
Now it has changed to: price increase + priority scheduling + profit optimization
4. Who will benefit?
·GPU/chip manufacturers (demand continues to explode)
·Cloud service providers (profit improvement)
·Leading AI applications (can access resources)
Who will be squeezed? Small and medium-sized AI startups (cannot access computing power or costs rise)
USDC issuer Circle (CRCL) has just done something "far-reaching"
Appointing Microsoft executive Kirk Koenigsbauer to the board, while also joining the Compensation Committee + Risk Committee.
Who is he?
· Current President and COO of Microsoft’s Experience and Devices division
· Responsible for Microsoft 365 and Microsoft Copilot
· Has worked at Microsoft for over 30 years, personally driving Office 365 to the cloud
More crucial than the "appointment" is the signal
1. Circle is fully aligning with "global enterprise-level financial infrastructure"
Not Crypto Native, but: global enterprise services + compliance + risk control system
2. Board structure upgrade = preparing for bigger moves
The entry of such high-level individuals into the board usually indicates: large-scale institutional collaborations, global expansion
3. Why the Microsoft connection?
Because the future competition for stablecoins is fundamentally about: enterprise-level payments + cloud + AI + security systems
Appointing Microsoft executive Kirk Koenigsbauer to the board, while also joining the Compensation Committee + Risk Committee.
Who is he?
· Current President and COO of Microsoft’s Experience and Devices division
· Responsible for Microsoft 365 and Microsoft Copilot
· Has worked at Microsoft for over 30 years, personally driving Office 365 to the cloud
More crucial than the "appointment" is the signal
1. Circle is fully aligning with "global enterprise-level financial infrastructure"
Not Crypto Native, but: global enterprise services + compliance + risk control system
2. Board structure upgrade = preparing for bigger moves
The entry of such high-level individuals into the board usually indicates: large-scale institutional collaborations, global expansion
3. Why the Microsoft connection?
Because the future competition for stablecoins is fundamentally about: enterprise-level payments + cloud + AI + security systems
Login to explore more contents