Cryptopolitan

Arthur Hayes believes US Treasury Secretary Scott Bessent is preparing a way to strengthen the yen that could also force the Federal Reserve to create a large amount of new dollar liquidity. Arthur’s case centers on the Fed’s FIMA Repo Facility.
In his latest essay titled ‘Yen-Quake,’ Arthur said that Japan could place its US Treasury holdings with the Fed as collateral, borrow dollars, sell those dollars for yen, and then use the yen to buy Japanese assets.
However, what Arthur perceives to be important about the setup is the creation of dollars because a larger balance sheet at the Fed may feed directly into Bitcoin, Ether, gold, miners, and other financial assets.
Arthur says rate hikes and foreign asset sales would create pain Japan and the US want to avoid
Arthur lays out three ways Japan could strengthen the yen. The BOJ could raise rates aggressively. The government could push GPIF and other institutions to sell overseas assets and bring the cash back home.
Or the Ministry of Finance could borrow dollars from the Fed through FIMA and use those dollars to buy yen. Arthur believes the third choice is the one Washington and Tokyo can live with.
The interest rate gap makes choice one harder to implement. Arthur mentions that dollars currently yield some 2.75 percentage points more than yen. Traders can borrow yen at low rates and then swap them into dollars to purchase Treasury bills. Increasing interest rates in Japan will decrease this margin and support the yen.
On the other hand, the Bank of Japan holds a huge amount of Japanese government bonds due to the years of yield curve control policy. Increasing interest rates will make their bond holdings less valuable and create even greater unrealized losses for the central bank. Moreover, Japan’s government will have higher expenses as a result of increased yields on its debt.
“What happens when rates rise? Bond prices fall. The lower bond prices fall, the larger the BOJ’s unrealized loss. Unlike you readers, the BOJ can lose an infinite amount of yen because it can print them at will.”
Arthur also points out that a sudden jump in the yen can force traders who borrowed the currency to finance stocks and bonds elsewhere to close those positions quickly.
The second choice is politically cleaner inside Japan but far more dangerous for US markets. GPIF manages roughly $1 trillion to $2 trillion. Its 2014 allocation changes increased purchases of foreign stocks and bonds, creating a steady source of yen selling.
Arthur says Japanese officials are now discussing putting more money into domestic securities. If that policy eventually reaches GPIF, hundreds of billions of dollars could come back to Japan.
But also, if Japan becomes a major seller of US Treasuries and American stocks, Washington gets a stronger yen but also loses one of its biggest foreign sources of demand for US assets.
Arthur says neither country wants to find out what happens if USD/JPY falls from around 160 toward roughly 90, which he cited as a purchasing-power estimate of fair value.
Bessent wants FIMA to let Japan buy yen while the Fed supplies the dollars
Arthur says the third choice avoids forcing Japan to sell its Treasury holdings, with the Ministry of Finance pledging Treasury securities through FIMA. The Fed then lends dollars against that collateral. Tokyo sells those dollars in the currency market and buys yen. The yen can then go into Japanese government bonds and domestic stocks.
“MOF repos a treasury security and receives a dollar loan from the FIMA program at the Fed. MOF sells dollars and buys yen in the global forex market. MOF reinvests the yen domestically by purchasing JGBs and stocks.”
Arthur argues the Fed would have to create the dollars used for those loans. As FIMA borrowing grows, the central bank’s balance sheet would grow with it.
That is where his Bitcoin thesis starts. The collateral pool is huge. Arthur estimates that Japan’s government owns $1.143 trillion of US Treasuries. GPIF holds another $230 billion.
Together, that gives them about $1.373 trillion in Treasury securities.
For context, the Fed’s balance sheet expanded by roughly $4 trillion between 2020 and the end of 2021.
Arthur says the Fed’s Foreign Currency Subcommittee can alter FIMA’s operating rules. He expects Trump and Bessent to push Fed Chair Kevin Warsh toward the changes they need.
Arthur finished his essay saying that his existing positions are already centered on Bitcoin, physical gold, and gold miners.
Arthur also sees Ether as the major large-cap crypto alternative because “the narrative is the one major shitcoin that didn’t eclipse its all-time high in 2025; in addition, Ethereum will be the security layer for RWAs.”
Don’t just read crypto news. Understand it. Subscribe to our newsletter. It's free.

Elon Musk gave a one-word reply when he saw the news of another high-level executive leaving OpenAI. Barely 24 hours passed between Musk’s “Wow” reaction to the news that Chloe Bakalar, former head of ethics, was leaving before Brad Lightcap, one of OpenAI’s longest-serving executives, announced his departure.
Musk has had public spats with the CEO of OpenAI, Sam Altman, since he left the company in 2018. His reply garnered over 12,000 likes by Tuesday. Musk did not expatiate, as the brief reply drew lots of comments.
Why did Lightcap leave OpenAI?
Lightcap informed OpenAI staff of his departure on Tuesday. He reposted his goodbye note on X.
Lightcap served as chief financial officer and chief operating officer during his eight years at the company. He turned his attention to a special projects team in April, reporting directly to Sam Altman. Lightcap described his decision as “bittersweet,” stating he was going “to start something new” without elaborating on his plans.
His stint with OpenAI began in 2018, joining the fledgling AI startup after working at Y Combinator, Dropbox, and J.P. Morgan. Lightcap intends to remain at the company for the next couple of weeks.
OpenAI has not commented so far.
A litany of exits at the top
Lightcap’s exit is not an isolated incident. It comes amid a series of exits at OpenAI. Fidji Simo, who is considered OpenAI’s number-two executive, resigned in July due to health concerns. He’s gone on to join an AI health startup.
The man who ran OpenAI’s Safety Systems team, Johannes Heidecke, stepped down from the company after changes in the structure of the safety and research teams.
Joshua Achiam, who served as OpenAI’s chief futurist, left after nine years in July without stating any specific reason. He vowed to continue the mission “from outside the walls of a frontier lab.”
This is a pattern that goes way back. Since 2024, a long list of safety and alignment leaders have departed OpenAI. This includes co-founders Ilya Sutskever and Jan Leike and the head of ethics at OpenAI, who joined within the past year.
Why is Musk talking about OpenAI exits ahead of IPO?
The timing adds more meat to the story. The departures come as OpenAI looks toward an IPO.
Typically, investors are secure in the knowledge that a company has stability in its executive team. But the recent events might do the exact opposite. However, the exits could also be explained as part of a growing trend in the AI industry. Recently, Jeff Dean exited Google to start up Discovery Loop, his own AI venture.
As for Altman and OpenAI, the AI lab could have done without this many executive-level departures as it ramps up to its IPO. Musk’s reaction is the latest chapter in the rivalry between him and OpenAI’s Altman, as he builds his own AI company.
Don’t just read crypto news. Understand it. Subscribe to our newsletter. It's free.

ENS token holders have ratified the “Next Era of ENS DAO” proposal and pushed the executable onchain, standing up the ENS Foundation as a working organization with a paid executive director, staff, and a board of five, ENS Labs said in an August 11 blog post.
The change gives ENS its first legal body able to sit in rooms where a DAO cannot. Also, the body will be handling the day-to-day operations while the token voters will continue to control the protocol.
Why a DAO couldn’t sign the paperwork
According to ENS Labs, a DAO has no standing to sign standards agreements, employ full-time people, take a trademark holder to court over a phishing clone, or answer legal process.
Before now, most of the representative work went undone, with ENS Labs handling some of it, a role that they were not well-equipped to speak on, given that they are primarily focused on engineering.
Now that the Foundation is live, ENS Labs says that it can now focus on building, and this includes the ENSv2 upgrade.
The Foundation is tasked with carrying the offchain load. It can now join ICANN, IETF, and W3C discussions and chase recognition of the .ens top-level domain.
As a named entity, it is also responsible for meeting with regulators, registrars, standard bodies, and courts. It is also tasked with consolidating the ENS trademarks and brand assets so it can move against impersonators.
What does ENS say about where the money stays?
ENS DAO’s holdings is 54.6% of the total token supply, and it will continue to stay under the control of the DAO, which comprises tokenholders.
The Endowment, funded by .eth registration revenue, keeps its assets where they are; however, its transactions now run through a nine-day timelock.
During this nine-day window, the ENS Security Council can terminate any transaction that falls outside the Foundation’s mandate.
In the same post, it was highlighted that the ENS Foundation and ENS Labs are not the same legally, with each one operating as a separate entity. The Foundation does not own ENS Labs, and the latter has no governance rights over the Foundation, the endowment, or the protocol.
The closest relationship the lab has with the Foundation is that the founder has a founder seat on the Foundation’s 5-person board; however, they are not involved in the decisions that pertain to funding for the Lab.
Who are the members of the ENS Foundation five-seat board?
The inaugural board includes Alexander Urbelis, a cybersecurity lawyer who has served as general counsel and Chief Information Security Officer (CISO) at ENS Labs and CISO of the NFL, who takes the executive director role.
ENS founder and Labs CEO Nick Johnson holds the Founder seat.
Kartik Talwar of A.Capital Ventures and ETHGlobal, Brett Sun, the cofounder of Prelude, and Anthony Leutenegger, CEO of Aragon, make up the remaining three independent directors.
The independent directors will be serving a two-year term, which can be renewed by the tokenholders.
The update from ENS comes in a year that has already seen some governance scuffles on the platform. In June, a temp check proposal was published with the aim of restructuring their operations with a focus on the DAO and the foundation.
Johnson, the CEO who holds an estimated 3.26 million ENS, has also been the point of some controversy as he has the ability to change the direction of votes given the size of his holdings. This came up when he was reportedly able to drive about 80% of the votes cast against a Security Council renewal in late June.
They later approved a successor eight-member council with a stricter 5-of-8 multisig for a two-year term.
ENS currently trades around $4.12, having declined by over 1.9% in the past 24 hours, per CoinMarketCap data. It is now below the $85.69 peak it hit in November 2021 by more than 95%.
If you're reading this, you’re already ahead. Stay there with our newsletter.

Cole Villemain, the Pudgy Penguins co-founder who holders voted out in 2022 over accusations that the team drained the project’s treasury, has sold out a 44,444-piece NFT collection on Robinhood Chain.
On-chain observer @0xmani reports that the collection was sold for more than 684 ETH (roughly $1.28 million).
Is Cole Villemain running an NFT scam?
Cole Villemain, who posts as ColeThereum on X, announced that he has sold out his 44,444-piece NFT collection called Spritehood Wisps with a two-line message reading: “Sold out. I’m back in the NFT game.”
Some traders, like @CryptoGorilla, pointed out that the mint was promoted as a “free mint” but ended up being a “stealth launch” that cost money to join. Others, like user HanIsRich, called the project a “cashgrab,” arguing that the promised future development wouldn’t cost nearly 20% of the money raised.
Gorilla stopped short of calling it a scam, noting that the floor had climbed 2.6 times off its minimum and that rare pieces were changing hands as high as 0.88 ETH.
The collection was listed under the creator name Spritehoodio, and traded on OpenSea with a floor price of $35.21 as of publication.
Spritehood Wisps on OpenSea. Source: OpenSea.
OpenSea data shows 4,074 unique owners across the 44,444 tokens, meaning a single holder controls roughly 9 out of every 10 pieces on average. Only 5.2% of the supply is listed for sale, and 24-hour volume stood at $781,000 with a top collection offer of $32.42.
Villemain built the drop on Robinhood Chain rather than Ethereum. The brokerage launched that network on July 1, 2026, as a place for tokenized stocks and U.S. Treasuries, but memecoins overran it instead.
In its first few weeks, Robinhood Chain documented wallet drainers, phishing pages, and rug-pulls spreading across it.
What was the Pudgy Penguins co-founder accused of?
In August 2021, blockchain investigator ZachXBT profiled Villemain’s pre-crypto dropshipping site, eBoy Outlet, describing its reviews as full of customers who never got orders, refunds, or support responses. Villemain denied wrongdoing and said he refunded buyers who did not receive goods.
Villemain also launched a prior NFT collection, My Fucking Pickle. Its floor now sits at $13, down 98% from a June 2021 high above $540.
“Have to love cash grab projects,” ZachXBT wrote at the time.
Reacting to the new drop, one developer wrote on X saying that Cole was the “same guy who did early meme NFT cash grabs” and that “half of Crypto Twitter is acting like none of that ever happened.”
Pudgy Penguins, which Villemain cofounded, launched in July 2021 as a set of 8,888 cartoon penguin avatars on Ethereum, minting at 0.03 ETH and going on to total $145 million in OpenSea volume.
On January 5, 2022, an investor alleged that the founders had emptied the collection’s ETH treasury. Holders voted the founders out through a Discord poll the next day, and Villemain stepped back from X, citing “mental health.”
By April 2022, the individuals left in the team sold Pudgy Penguins to a group led by Los Angeles entrepreneur Luca Netz for 750 ETH, worth about $2.5 million then. Netz later turned the brand into one for physical toys that have moved more than a million units through Walmart, Target, and Walgreens.
Don’t just read crypto news. Understand it. Subscribe to our newsletter. It's free.

The new CEO of Twenty One Capital (NYSE: XXI) Raphael Zagury has announced to shareholders that the Bitcoin treasury firm must grow beyond its balance sheet.
He made the announcement on Tuesday, even as he listed out five main objectives for the Tether-controlled firm following a $1.27 billion loss in H1 of 2026.
Bitcoin’s slump caused most of Twenty One Capital’s loss
The loss of $1.27 billion was linked to movement in the market rather than an operational failure. The major cause was the crash in the price of Bitcoin, which makes up most of the company’s portfolio.
Twenty One Capital, based in Austin, Texas, trades on the NYSE using the XXI ticker symbol. As of June 30, the company holds 43,514 BTC, which is worth ~$2.8 billion right now.
Twenty One Capital is currently the second-largest corporate Bitcoin holder. The company disclosed the purchase of 6,284 BTC in July, while Michael Saylor’s Strategy, the largest corporate Bitcoin holder, has sold ~7,000 BTC this year.
XXI filed its quarterly report for Q2 with the SEC on Tuesday, August 11. And its revenue is practically zero, showing the need for an operating plan more than ever.
Zagury names five priorities
Zagury intends to transform XXI into more than just a large Bitcoin holding alone because he believes the company must do more to convince traders to buy its stock.
“Twenty One owns one of the largest Bitcoin balance sheets in the public markets. That is a real advantage, but if Twenty One is going to be worth owning, it must become more than a Bitcoin treasury,” he said in his letter to shareholders.
He listed five areas of focus.
First, tighten governance, controls and reporting and hire the staff to run the strategy.
Second, buy and build operating businesses.
Third, develop new capital markets products.
Fourth, widen the firm’s ability to do mergers and acquisitions.
Fifth, stand up a Bitcoin-backed lending and credit arm, which Zagury said could later manage outside capital if kept conservative and low-leverage.
Zagury also committed the firm to backing Bitcoin’s open-source infrastructure and to watching transactions with Tether, its controlling shareholder, closely.
Mallers out, merger plan scaled back
Zagury took the top job after Jack Mallers stepped down, effective July 20, to return to Strike, the Bitcoin payments company he founded. Mallers helped build Twenty One and steer it to a public listing in December 2025.
His exit came with a strategy reset. Tether had proposed in April to fold Twenty One together with Strike and mining firm Elektron Energy into one listed Bitcoin business spanning treasury, payments and mining.
That three-way deal has been abandoned, with Strike dropping out. The company is now weighing a narrower two-way combination with Elektron. Zagury previously ran Elektron and has held senior roles at Goldman Sachs, Deutsche Bank and Merrill Lynch.
Why the market remains cautious
The reset lands with the stock near its lows. XXI closed at $4.58 on August 10, the SEC filing states, down from a 52-week high of $12.51 and off more than half its value this year.
Tether tightened its grip in May by buying SoftBank’s entire stake for about $711 million, removing the last outside sponsor from the board.
For shareholders, the message is that patience is now the ask. Investors should watch whether the promised operating businesses, lending arm and any Elektron deal actually materialize, and whether the governance overhaul produces the reporting discipline a company with a multibillion-dollar treasury and three employees currently lacks.
Don’t just read crypto news. Understand it. Subscribe to our newsletter. It's free.

The Central Bank of Russia has greenlighted Bitcoin, Ethereum, and Tether’s stablecoin for trading on licensed cryptocurrency exchanges.
The leading coin will be admitted to the regulated Russian market for digital assets as part of new purchase rules for non-professional investors.
Russia’s central bank whitelists BTC, ETH, and USDT
The monetary authority in Moscow has included the top three coins by market cap in a list of cryptocurrencies that will be available to ordinary Russians.
“Non-qualified” investors in the country will soon be able to acquire Bitcoin (BTC), Ethereum (ETH), and Tether’s dollar-pegged stablecoin USDT, according to a draft directive.
The document has been published online for public discussion, and the regulator will accept comments and suggestions by August 24, Russian media reports revealed on Tuesday.
It details what the largest category of market participants can and cannot do under the new law “On Digital Currencies and Digital Rights,” which will regulate crypto transactions in Russia next month.
The long-awaited legislation, which was passed by both houses of parliament in the second half of July and signed by President Putin in early August, is scheduled to enter into force on September 1.
Meanwhile, the Central Bank of Russia (CBR) published several additional regulations for various operations with cryptocurrencies, such as exchange, storage and margin trading.
According to an earlier statement by its First Deputy Chairman Vladimir Chistyukhin, the monetary authority will have to prepare more than 30 such directives by November.
These will introduce specific rules to govern the activities of traditional and fintech market players, including stock exchanges and new entities such as “digital depositaries,” or crypto custodians.
As per the latest directive, non-professional investors will be able to buy the three cryptocurrencies from any CBR-approved intermediary, including brokers, exchanges, and asset managers.
However, their purchases will be capped, and the annual limit has now been set at the previously discussed 300,000-ruble threshold, or a little over $3,600 at the current exchange rate.
In its announcement published on August 11, the financial authority emphasized:
“To protect non-qualified investors from sharp and unpredictable fluctuations in cryptocurrency prices, only the most liquid cryptocurrencies will be available to them.”
Approved cryptocurrencies chosen based on strict criteria
Under the “digital currency” law, the coins are selected in accordance with concrete criteria regarding market capitalization, average daily trading volume, and pricing history on foreign exchanges.
As previously reported by Cryptopolitan, in order to be whitelisted in Russia, a crypto asset’s market cap must have exceeded an average of 5 trillion rubles (over $60 billion) over the past two years.
Its average daily trading volume should be above 1 trillion rubles for the same period (more than $12 billion), and it’s required to have a trading history of at least five years prior to admission.
In June, the CBR’s deputy governor, Chistyukhin, also indicated that only three cryptocurrencies meet these requirements currently. The institution now said in its press release:
“Based on these criteria, the Bank of Russia has included Bitcoin, Ethereum, and Tether USDT in the list of cryptocurrencies available for public trading on exchanges.”
The bank also pointed out that qualified investors will be able to acquire all cryptocurrencies traded on Russian exchanges and over-the-counter markets without restrictions.
At the same time, it’s reminded that before making any transactions, all investors, regardless of their status, will have to undergo testing to verify their awareness of the associated risks.
Regular Russian citizens, which are likely to form the majority of investors after the comprehensive legal framework comes into force, remain quite skeptical about the new opportunities it will offer.
A recently conducted survey showed that almost 70% of Russians do not expect to see any significant use cases for cryptocurrency after its legalization under the terms and limitations offered by their government.
The smartest crypto minds already read our newsletter. Want in? Join them.

BRICS members are exploring plans to connect their CBDCs and fast payment systems, a move that could cut the cost of moving money between the bloc’s economies, Reserve Bank of India Governor Sanjay Malhotra said on Tuesday.
Malhotra spoke at an event in Mumbai, where he described cross-border transfers as one of the priorities the group keeps returning to. He, however, cautioned that talks were at a very early stage. “Various options are on the table, but it is still at the discussion stage, including CBDCs (central bank digital currencies) and linkages of fast payment systems,” he said.
The governor stated that the appeal was in terms of cost saving: “Cross-border payments is an area of interest for all of us, including the BRICS, because we feel there is a lot of scope for reducing cost.”
Why BRICS linking CBDCs could cut costs
CBDCs are digital forms of countries’ official currencies, issued directly by their central banks. Creating a channel between these currencies could let payments travel more directly and with much more ease between nations. This will cut out layers of intermediaries that money usually passes through and lower what each transfer costs.
The route of payment with linked CBDCs will work differently, but aim at a faster way for consumers and businesses to send money abroad through the domestic pathways they already use. These paths will also widen the reach of the Indian rupee.
India aims to internationalize the rupee
RBI governor Malhotra said the central bank will keep working to internationalize the rupee and to promote local currencies for trade and payments across borders. India is hosting this year’s BRICS summit, an annual gathering that brings together five countries which include Brazil, Russia, India, China and South Africa.
The RBI had previously asked the Indian government to place a CBDC-linking proposal on the summit’s agenda earlier in 2026. The country’s goal to internationalize its currency is a major reason to continue to push to link CBDCs amongst the BRICS countries.
The governor also spoke at the event about Indian banks and financial institutions and their use of AI. He asked them to catalogue every AI model already running within their institutions and to adopt AI governance policies approved by their boards.
If you're reading this, you’re already ahead. Stay there with our newsletter.

Kazakhstan President Kassym-Jomart Tokayev has signed a decree exempting individuals from income tax on digital-asset gains for three years.
The decree is expected to move an estimated 1 million crypto wallets off foreign platforms and onto licensed domestic exchanges.
Three-year tax-free period in Kazakhstan
President Kassym-Jomart Tokayev has signed a decree created by three bodies; the Ministry of Artificial Intelligence and Digital Development, the National Bank of Kazakhstan, and the Astana International Financial Centre (AIFC), which states that private investors will owe no personal income tax on gains from digital-asset transactions for three years.
Assets linked to fraud, money laundering, or unlicensed crypto services are exempted from this decree.
Kazakhstan’s Vice Minister of AI and Digital Development, Gizzat Baitursynov, said his department is already drafting a simplified tax regime that will be implemented after the three-year window, and is separately working to cancel tax audits covering investors’ previous three years.
AIFC data says Kazakh citizens hold roughly 1 million crypto wallets, nearly four times the 256,900 users registered on authorized local exchanges as of March. Cryptopolitan has previously reported that some 95% of the country’s crypto turnover was changing hands outside the regulated market, in peer-to-peer deals or on foreign platforms.
In April, the Astana Financial Services Authority named HTX, Bitget, OKX, and MEXC as unlicensed operators, Cryptopolitan reported at the time.
How will Kazakhstan solve its electricity problem?
The decree also addresses the problem that broke Kazakhstan’s first mining boom: electricity. After China banned Bitcoin mining in 2021, the country became the second-largest mining hub after the United States. It was ranked third globally by hash rate in 2022, but that surge overwhelmed the country’s aging grid, and three power plants in the northeast shut down in an emergency in October 2021, triggering blackouts.
Miners at their peak drew an estimated 8% of national electricity output.
To keep new mining off the public grid, the order lets oil and gas fields divert associated petroleum gas the state does not need into autonomous generators for mining. A parallel “70/30” model gives data centers and miners direct access to up to 70% of new capacity built through infrastructure upgrades.
Nurkhat Kushimov, the general manager of Binance Kazakhstan, called the tax break the decree’s most important measure and said it makes licensed jurisdictions more attractive.
Bakhytzhan Kenzhebayev, who chairs Kazakhstan’s Association of Fintech, AI and Crypto Industry, said the exemption removes a key uncertainty for investors. However, he warned that loose legal definitions could invite abuse and force a reversal within a year or two.
Separately, the OECD’s Global Forum said Kazakhstan is implementing the Crypto-Asset Reporting Framework ahead of its first automatic exchanges of crypto tax data in 2027.
If you're reading this, you’re already ahead. Stay there with our newsletter.

IBM has committed to a multi-year, $240 million agreement with Together AI to build a dedicated Nvidia inference cluster on IBM Cloud, the two companies announced on Tuesday.
The new deal hands the open-source AI provider a block of enterprise-grade GPU capacity as it pushes deeper into large corporate accounts.
HGX B300 systems to hit IBM Cloud in early 2027
According to the signed agreement, IBM is set to place a large cluster of Nvidia HGX B300 systems inside IBM Cloud, with the hardware expected to come online in the first quarter of 2027, according to IBM’s newsroom release.
Together AI will use this capacity to serve inference on open-source models. IBM has claimed that it is the first dedicated cluster of this size built for inference on IBM Cloud around the B300, wired with Nvidia’s Spectrum-X Ethernet networking.
Nvidia stated that the configuration can generate 30 times more “AI factory output” than its previous generation, although this claim is not independently verified.
Together AI’s 400 trillion monthly tokens to run on IBM
Together AI had an $800 million Series C funding raise in recent months at an $8.3 billion valuation. This was aimed at expanding what it calls its AI Native Cloud.
The AI company, founded in 2022, rents cloud infrastructure to companies that want to build on open, modular AI stacks instead of closed models. IBM’s press release also mentioned that the company now channels 400 trillion tokens a month through its inference product and helps to power more than a million developers.
The IBM cluster gives Together AI’s planned expansion a platform to run on. CEO Vipul Ved Prakash said the deal was a way to reach more customers without forcing them to pay frontier-model prices.
“Enterprises want the performance of the best frontier models without the closed-model price tag, and that only works if the infrastructure underneath is fast and reliable at scale,” he said in the release. He also added that the cluster is “a big step in our push to make open-source AI the obvious choice for enterprises.”
IBM furthers Nvidia partnership to court AI startups
For IBM, the deal with Together AI is another entry in an organically increasing partnership with Nvidia. Alan Peacock, general manager of IBM Cloud, explained that the deployment was tied to corporate demand for agentic AI, saying IBM and Nvidia are “delivering scalable, economical, enterprise-grade AI infrastructure” to help Together AI move faster.
Dion Harris, a senior director for HPC and AI infrastructure at Nvidia, described AI factories as infrastructure that is becoming as basic as electricity or telecommunications.
IBM stated that the two companies have also been working together on GPU-native data analytics in addition to unstructured data extraction. The companies are also focused on consulting and positioning the Together AI cluster as an important part of a larger push to sell AI capacity to both large enterprises and startups.
The smartest crypto minds already read our newsletter. Want in? Join them.

eToro (NASDAQ: ETOR) beat Wall Street expectations for its business when it reported $53 million in Q2 net income as markets opened on Tuesday. The trading platform announced that it had reached a $231 million deal to acquire the US brokerage TradeZero in the same disclosure that also revealed a sharp drop in its crypto trading section from the year before.
The news of the estimate beat did not give the firm’s ETOR stock a much-needed lifebuoy as it continued the downtrend from the previous session. eToro is now down more than 8% in early Tuesday trading, exchanging at $31.15 per share as of this Cryptopolitan report.
eToro continued to slide after releasing Q2 results. Source: Google Finance.
The slide continues what has been a year to forget for eToro after seeing its market value drop to $2.51 billion per Google Finance.
A GAAP net income jump of 77%
Per the documentation from eToro, the firm posted adjusted diluted earnings of $0.68 per share for the quarter ended June 30 on its Nasdaq-listed ETOR ticker. Beating the $0.61 consensus extends eToro’s streak to four consecutive quarters.
What were the headline numbers from eToro’s Q2 earnings release?
On a GAAP basis, net income rose to $53 million from $30 million, a 77% year-on-year increase.
Equities trading helped net contribution reach $229 million, up 9%.
Funded accounts climbed 18% to 4.28 million.
Assets under administration (AUM) reached $19.2 billion.
Cash, cash equivalents and short-term investments were estimated at $1.2 billion.
CFO Meron Shani echoed a sentiment that many in the industry have shared during the quarter: customers are rotating across markets rather than any single asset class.
eToro users backed up that trend as 60% of them expanded into equities in Q2 after trading commodities between Q4 2025 and Q1 2026. Nearly 90% of that demographic also traded crypto during the same period.
Why is crypto volume down on eToro?
The decline in crypto trading volume and activity on eToro tracked an industry-wide slowdown. Total cryptocurrency trades dropped to 1.4 million in July 2026, down 73% from 2025. Traders were making smaller moves too, as the average amount invested per trade dropped 50% to $182.
The slowdown in eToro’s crypto business showed up on its balance sheet. Revenue fell roughly 30% to $1.34 billion from $1.9 billion a year earlier.
Equities and commodities trading have been getting a lot of the action that has left crypto trading, as they combined for $141 million in net income during a period where crypto assets contributed only $19.7 million.
Cryptopolitan reported in July that eToro’s crypto profit had shrunk to about 5% of net trading profit in Q1 2026.
Why TradeZero matters for the US push
The acquisition is the piece aimed at fixing eToro’s weakest geography. TradeZero, founded in 2015, runs online brokerage operations across the US, Canada and international markets, and serves a base of active traders.
eToro will pay cash plus up to 2.5 million newly issued Class A shares, with the deal capped at $231 million and expected to close in the first half of 2027, pending regulatory sign-off.
What eToro is really buying is broker-dealer infrastructure. That plumbing is harder to build than it is to acquire in the US, where launching regulated products takes more than porting over technology from another market.
TradeZero generated about $80 million in revenue in the 12 months to June 30 with an 81% gross margin, Calcalist reported, valuing it at roughly 2.9 times sales at the top of the range. Jefferies advised eToro, which called this its third acquisition signed in 2026.
“Today’s announcement is an important step in building our US business,” CEO Yoni Assia said. The company has been chipping away at that market for a while, having entered New York in April after a years-long wait for authorization under the state’s BitLicense regime.
The smartest crypto minds already read our newsletter. Want in? Join them.

By Pierre Gérard, CEO and co-founder, Scorechain
When we founded Scorechain in Luxembourg in 2015, “blockchain analytics” was not yet a category. We spent the first few years explaining to banks and regulators why the transparency of a public ledger was an opportunity rather than a threat. A decade later, I am watching the same misunderstanding attach itself to artificial intelligence (AI), and it is costing the industry time it does not have.
Two stories dominate the conversation. The first says AI will soon replace compliance teams altogether. The second says AI is too unpredictable to let anywhere near regulated financial activity. I do not believe either one, and I say that as someone whose company has risk-assessed more than 2,800 virtual asset service providers (VASPs) since 2015, and has spent the past two years adding AI where it genuinely helps, as a separate layer rather than something baked into the compliance tools our clients depend on.
Start with the problem that every compliance officer I speak to raises within the first five minutes: noise. A sanctions screening system tuned the way a nervous bank tunes it can throw off false positives on the order of 95%. Transaction monitoring is not far behind. So a trained analyst, someone who understands typologies and can read a fund flow, spends the bulk of the working day clearing alerts that were never risks: dismissing name matches on a common surname, reading five adverse media hits that turn out to describe a different person entirely. Each of those adverse media checks takes an analyst 10 to 20 minutes. That is the actual texture of compliance work today, and it is why good people burn out of the profession.
This is where automation earns its place, and it is a narrower place than the hype suggests. I am not neutral about it. Scorechain AI exists to hand an analyst a single report: a wallet’s risk score, the entity types it has interacted with, and the named services and counterparties it has been exposed to. That is work that used to mean hours of manual tracing across a ledger. But notice what the report does and does not do. It does not decide anything. It compresses the evidence so that the compliance officer, the person who has to sign off on that decision and defend it to a regulator later, can read it in minutes and then make the call. That is the whole game. Good automation does not shrink the compliance function; it moves it off the treadmill of triage and back toward judgment.
The distinction matters, because the alternative is dangerous. In a regulated setting, a model cannot answer to a supervisor. The Sixth Anti-Money Laundering Directive (AMLD6) and the Markets in Crypto-Assets Regulation (MiCA) both require an institution to explain and stand behind its decisions. “The algorithm flagged it” is not a defence at an inspection, and “the algorithm cleared it” is worse. So the only responsible design is AI as a support layer sitting on top of trustworthy data, with a named compliance officer retaining the decision and the accountability that comes with it. Human oversight is not a training-wheel we remove once the model matures. It is the architecture.
And a model is only ever as good as what sits beneath it. This is the part outsiders miss. On its own, an AI reading a blockchain sees only anonymous strings of characters moving value to other anonymous strings. It cannot tell that the wallet three hops upstream is a sanctioned exchange, or that the counterparty receiving the funds is a mixer rather than a payroll provider. Supplying that missing context is the whole job of blockchain analytics: attaching identity and risk to raw on-chain activity, tracing indirect exposure across multiple hops rather than just checking the address in front of you, and scoring it against the more than a billion data points and over a million crypto entities we have labelled since 2015. Take a concrete case: a wallet looks clean at first glance, but tracing its flows shows that most of its balance arrived, two hops back, from an address tied to a ransomware operator. That is the finding a model would never reach on raw chain data alone, and the one a compliance officer has to act on. Feed a model that context and it can reason on solid ground. Feed it thin data and it produces confident nonsense, which in compliance is more dangerous than an honest gap, because it clears things it should not.
Here is what I find genuinely new. AI is no longer only a tool that compliance teams use; it is becoming a participant in the market they monitor. Autonomous agents that initiate payments under preset limits have moved from demo to deployment, pushed along by real infrastructure: Coinbase’s x402 standard for machine-to-machine payments, Visa’s Trusted Agent Protocol, the PayPal and OpenAI checkout integration. Software is starting to transact with other software, settling in crypto assets, at a volume no treasury team could match by hand.
It raises a question the industry has not answered cleanly yet: how do you apply Know Your Transaction principles to a counterparty that is a piece of software? The direction, at least, is clear. When agents transact on their own, controls cannot live only at onboarding. They move to the transaction layer itself: real-time monitoring, velocity limits, provenance, and the ability to intervene while money is still in flight. The transparency we spent years defending to sceptics turns out to be the one thing that makes autonomous on-chain activity auditable at all.
This is the future we decided to build for rather than wait on. We recently launched Scorechain MCP, which exposes our risk scoring and entity intelligence through the Model Context Protocol, the emerging standard that lets AI agents call external tools directly. The intelligence lives in our platform, and the AI stays outside it, calling in for answers rather than being embedded in the compliance tool itself. The premise is simple: an agent should never transact blind. Before it moves funds or approves a counterparty, it can ask Scorechain in the same breath whether that address is a sanctioned entity, a mixer, a known scam, or a clean private wallet, and receive a risk score in return. This is not a hypothetical throughput. We already run more than 1.5 million AML checks a day, and a screening call returns in roughly 235 milliseconds, quick enough to sit inside a live transaction rather than slow it down. We put it where those agents and workflows actually live, as an app inside ChatGPT and Claude, and as an integration on automation platforms such as n8n and Zapier. A compliance check that sits inside the flow, at the moment the decision is made, is worth far more than one bolted on after the money has already moved. The most basic check of all, whether an address appears on a sanctions list, should not sit behind a paywall for anyone. That is why we offer it as a free sanctions screening API that any developer, agent, or workflow can call. Screening for sanctions exposure is not where a compliance provider should be extracting value; it is the floor the whole market should be standing on.
There is a second-order shift here that token issuers and asset managers are only starting to reckon with. When value moves into stablecoins and tokenised assets at machine speed, the risk that matters is no longer only the individual transaction but the asset itself: who holds it, how concentrated that ownership is, and how much of the supply sits with sanctioned or otherwise high-risk entities. That is a different question from transaction monitoring, and it is the one our Digital Asset Intelligence is built to answer, giving an issuer or an asset manager an asset-level view of holders and exposure before they mint, list, or allocate.
Europe is readier for this than it is given credit for. MiCA and AMLD6 already assume continuous monitoring and clear accountability rather than a one-time check at the door, and a regime that assumes activity must be explainable is exactly what you want when software starts moving money. So yes, AI is coming for crypto compliance. It will remove a great deal of tedious work, and I welcome that. What it will not remove is the need for judgment, accountability, and verifiable data. It raises the bar on all three. The teams that treat AI as a faster analyst, grounded in reliable on-chain intelligence and kept firmly under human control, are the ones who will still be standing when the machines start transacting. That is closer than most people think.

Sony Group and TSMC signed a binding agreement on Tuesday to build a $4.69 billion joint venture in Kumamoto, Japan, to make next-generation smartphone image sensors in a deal that helps to track where the world’s camera chips get built and how much Taiwan’s top chipmaker is betting on Japan.
The venture will be called Advanced Vision Semiconductor Manufacturing Corp., TSMC said in its statement, and it will be located in Koshi City, in Kumamoto Prefecture.
Sony takes charge of new venture
Sony’s semiconductor arm, Sony Semiconductor Solutions, will keep the controlling stake in the venture. The arm will run the business as a subsidiary of Sony Group, and a director will be appointed from Sony’s side of the venture.
The funding is also split unevenly, with Sony Semiconductor Solutions putting in about 465 billion yen, equal to approximately $2.92 billion in part cash and part assets moved over through a company split. TSMC’s investment share is said to be about 282 billion yen. Nippon.com, citing Jiji Press, put the combined figure at 747 billion yen, which comes to roughly $4.69 billion according to the 159.33 yen-per-dollar cited by Reuters.
The Taiwanese chipmaker stated that the investment contributions will come in stages, and will be timed to market demand and business conditions. This suggests both partners will have the chance to slow down on the funding if orders reduce.
TSMC also stated that the investment assumes that the venture receives sufficient backing from the Japanese government, with the partners planning to take that question to the Ministry of Economy, Trade and Industry. Tokyo has already spent heavily to pull chip manufacturing into Japan’s borders, and this joint venture has been made expecting more of the same.
Sony focuses on designs, TSMC manufactures
Sony Semiconductor Solutions will lead the core sensor technology, product planning and design, while the venture leans on TSMC’s advanced-process technology and manufacturing experience to pull these designs into mass production. The two companies have said they “intend to accelerate the commercialization of image sensor products driven by customer needs,” according to TSMC.
The venture’s commercial production is set to fully commence in 2029. The plan integrates new development and production capacity into a Sony image sensor plant already built in Koshi, and the venture is not only aimed at the smartphone market, but the AI, automotive and mobile markets in addition.
Kumamoto’s history with TSMC
TSMC’s first advanced wafer plant in Kumamoto, run by its JASM unit, started commercial operations at the end of 2024 using specialty processes, and a second fab is currently under construction with 3-nanometer mass production expected in 2028.
The Kumamoto prefecture already has a growing local supply base and the joint venture with Sony being developed here places it close to already existing TSMC capacity.
The companies first floated the plan in May through a nonbinding memorandum of understanding, and the signing of the deal after a two-day meeting converts that intention into a legally binding definitive agreement.
Don’t just read crypto news. Understand it. Subscribe to our newsletter. It's free.

Bitcoin is steeply lagging behind equities, and until it begins to regain strength, the current market will remain dominated by stocks, according to Glassnode.
Over the past 90 days, Bitcoin has dropped by 20%, four times as much as the S&P 500, which rose by 5% in the same period.
Looking further back, the crypto market has performed the worst among major asset classes so far this year.
Bitcoin is down 35%, and altcoins are down 57% since January.
In contrast, gold is up 60%, copper 66%, and silver 107%. Nasdaq and Russell 2000 see a 38% and 31%, respectively, making crypto the weakest in the list over that period.
In a post on Tuesday, Glassnode said the market will remain led by equities until this gap closes.
Markets remain “equity-led tape” except Bitcoin moves, says Glassnode
Glassnode said this pattern continued over the past week.
“The last 7 days show the same pattern,” Glassnode wrote. “Until Bitcoin reclaims strength against the indices, this remains an equity-led tape.”
Over the last 90 days the S&P 500 rose 5% while Bitcoin fell 20%. The last 7 days show the same pattern.
Until Bitcoin reclaims strength against the indices, this remains an equity-led tape.
Notably, the NASDAQ is lagging behind other major indices indicating an unfavorable… pic.twitter.com/TpfDwoYB9y
— glassnode (@glassnode) August 11, 2026
Glassnode’s statement is evident in the Bitcoin market performance in July.
Artificial intelligence and semiconductor stocks suffered a steep market correction in July, with chip ETFs plunging over 20%. The Nasdaq-100 also fell nearly 7%.
However, Bitcoin and Ethereum gained 9% and 20%, respectively, in the same month, diverging from the rest of the stock market.
This is a big change from their previous close connection. In May, Bitcoin’s 90-day correlation with the Nasdaq was 0.89, according to TradingView data. By July 28, K33 Research found the 30-day correlation had dropped to 0.43.
BlackRock says the decoupling is “healthy”
Robert Mitchnick, who leads digital assets at BlackRock, said Monday that Bitcoin decoupling from stocks is “healthy” for investors who see Bitcoin as a way to diversify, calling July results a turning point for the crypto market.
Mitchnick also noted that crypto sentiment has begun to improve in the past month, with inflows returning to ETFs.
As of August 7, BlackRock’s iShares Bitcoin Trust (IBIT) had $48.51 billion in net assets, according to SoSoValue, and brought in $693.64 million in August.
“We have seen sentiment turn in a noticeable but subtle way the last month or so,” Mitchnick said.
US spot Bitcoin ETFs brought in $853.5 million in the week ending August 7, their best week since mid-April. IBIT made up about $693 million of that total.
Don’t just read crypto news. Understand it. Subscribe to our newsletter. It's free.

On August 11, 2026, Intel launched a sale of $20 billion worth of common stock, making it one of the largest stock releases by a chip manufacturer during the current AI spending fever. The transaction reveals the amount of capital that companies think they must raise to meet the increasing demand for AI computing.
This is also significant for cryptocurrency investors. Bitcoin miners have been converting their power capacity and data centers to AI and high-performance computing technologies, which means that they are competing for the same resources and investment as the companies engaged in these fields.
Intel sold 210,526,315 shares at $95 each, according to its press release, raising the deal from the $15 billion it had initially proposed. Underwriters also have a 30-day option to buy up to another 31,578,947 shares at the same price.
Intel expects net proceeds of about $19.7 billion and plans to use the money for general corporate purposes, capital expenditures and working capital. The offering is scheduled to close on August 12.
Why the timing lines up with a trillion-dollar spending year
The sale of stock occurs at the time when the investment in AI infrastructure is at an all-time high. According to a Goldman Sachs Research report, the total global investment in AI is likely to cross $1 trillion in 2026, where out of which almost $581 billion will be in the US alone. As per Joseph Briggs, an economist at Goldman Sachs, the total investment in AI can be $1.8 trillion by the end of this year.
Intel is using equity markets instead of incurring additional debt to support the endeavor. Priced at $95 a share, this offering far exceeds the September 2025 cost of $23.28 per share that NVIDIA agreed to pay as part of a $5 billion investment into Intel. That was a part of a larger arrangement wherein Intel was to develop NVIDIA-linked x86 CPUs using NVLink.
The massive difference in the prices indicates the extent to which the perception of the investors about Intel has changed as the firm expands its capabilities in the area of artificial intelligence and chip making.
A recovering balance sheet, but a foundry still fighting for share
Intel is gathering funds after experiencing significant growth in revenue. Revenue in the second quarter of 2026 amounted to $16.1 billion, which is an increase of 25% from the year before. According to CEO Lip-Bu Tan, this is the best revenue growth that Intel has ever enjoyed in more than 15 years.
Its Data Center and AI business grew 59% year over year, while Intel Foundry revenue increased 31%. The company nevertheless reported a GAAP loss of $2.16 per share, although non-GAAP earnings came in at $0.42 per share.
Intel still faces a steep manufacturing challenge. Counterpoint Research estimated that TSMC controlled 73% of the pure-play foundry market in the first quarter of 2026, compared with 7% for Samsung. Intel was not among the top five.
Closing that gap will require heavy investment in leading-edge manufacturing, including Intel 18A, which the company says is now in high-volume production in the US. Raising equity gives Intel more room to fund that expansion without taking on additional leverage.
Miners are already in the same fight for compute
The deal also has implications for Bitcoin miners, many of whom are chasing AI infrastructure opportunities. CoinShares said in its Q1 2026 mining report that Bitcoin miners had signed more than $70 billion in cumulative AI and high-performance-computing contracts. It also estimated that AI could account for as much as 70% of listed miners’ revenue by year-end, up from about 30%.
Companies such as Core Scientific, TeraWulf, Cipher and IREN are increasingly positioning themselves as data-center operators alongside their Bitcoin mining businesses.
The economics help explain the shift. CoinShares estimates mining infrastructure costs roughly $700,000 to $1 million per megawatt, compared with $8 million to $15 million per megawatt for AI infrastructure. That premium gives miners with access to power and data-center capacity an incentive to repurpose their assets.
Intel’s $20 billion raise adds another major source of capital to the race for chips, power, and advanced manufacturing capacity that now links AI companies, hyperscalers, and crypto miners.
What’s next?
The next important event is the closing on August 12, after which it will be seen if the underwriters opt for the 31.6 million shares option, which will mean the deal is now more than $20 billion in value.
J.P. Morgan, Goldman Sachs, Morgan Stanley, and Citigroup are the lead underwriters. Above the fundraising aspect, the question lies in execution. One still needs to see whether Intel can bring the extra funding into sufficient foundry and cutting-edge capacity to be able to compete in the AI market, which is expected to continue growing until 2028, according to Goldman.

The smartest crypto minds already read our newsletter. Want in? Join them.

Santiment data shows that the total number of Bitcoin wallets registered with a minimum balance of 10,000 BTC has reached 90, the highest level in 6 months, even as the smallest wallets have continued selling. Bitcoin (BTC) is trading at around $64,322 at the press time.
The divergence might provide a hint about the next market direction. Large investors are buying Bitcoin while smaller traders are withdrawing from it. According to Santiment, this pattern occurs before major price shifts in the market. Thus, the probability of Bitcoin crossing the $70,000 mark is higher than it moving below $60,000.
Six new elite wallets in eight weeks
Six wallets have crossed the 10,000-BTC threshold in the past eight weeks, lifting the size of that elite group by 7.1%, Santiment data show.
🐳 Bitcoin’s elite wallet count has bounced back to a 6-month high, with 90 wallets now holding at least 10K $BTC.
📈 This is not a tiny signal. There has been a net gain of +6 wallets holding at least 10K+ BTC in the past 8 weeks, a +7.1% rise.
🦐 Micro wallet holdings have… pic.twitter.com/c2khTjHsC5
— Santiment Intelligence (@SantimentData) August 10, 2026
Accumulation is also spreading further down the ownership ladder. Wallets holding between 10 and 10,000 BTC — a group commonly described as whales and sharks — have added about $1.5 billion worth of Bitcoin since July 29. At the same time, so-called micro-wallets have been shrinking throughout August.
The discrepancy arises in relation to two shocks that appear to have affected the smaller investors more than the biggest.
A hardware wallet hack and a crypto bill that did not pass
The first shock is related to the Coldcard exploit. Due to a firmware vulnerability that emerged in March 2021, wallets impacted by the vulnerability were deriving the seed using a weak random number generator in the software rather than the entropy chip contained in the device. Therefore, the keys for those who have been affected by the problem had a key strength of about 40 bits, according to information obtained from the TFTC (Terrorist Financing Targeting Center).
Up until early August, TRM Labs reported a loss of $116 million. However, since then, other sources have put the loss figures above $130 million.
The second shock originated from the United States government. The Senate delayed the vote on the CLARITY Act, which aimed at defining the roles of the SEC and CFTC in the regulation of digital assets, until September following the Democratic Party’s opposition to the bill.
John Thune, the US Senate Majority Leader, said, “The Dems are insistent on no Clarity vote.” He then went on to mention that proceedings concerning the bill were “queued up first thing when we come back.” Digital Chamber CEO, Cody Carbone, said the lack of progress on the bill “isn’t the result any of us hoped for” but added that “the fight is far from over.”
Neither event represents a flaw in Bitcoin itself. The Coldcard vulnerability was specific to one manufacturer’s firmware, while the CLARITY delay is a political and legislative issue. But both may have contributed to weaker holders moving coins toward wallets that appear more willing to hold.
New whales or the same custodians
An increase in the whale tally does not indicate the emergence of a new buyer class.
As Cryptopolitan previously reported, addresses concentrated around the 10,000-BTC level can be related to rearrangements in how large custodians store their balances rather than new capital entering Bitcoin. According to Arkham Intelligence, the largest Bitcoin addresses are usually owned by exchanges, custodians, and ETF issuers. Coinbase, for instance, has control over about 5% of the supply, whereas BlackRock’s ETF is estimated to hold 732,000 BTC.
Thus, transfers within one issuer from its cold wallets may result in the rise in large addresses without being associated with a new buyer.
The accumulating process is not something new. In January, Glassnode found out that wallets with a balance above 1,000 BTC increased from 1,207 in October to 1,303 due to retail selling amid a correction. However, Santiment found out in late July that wallets with a balance ranging between 10 and 10,000 BTC had sold off approximately 70,848 BTC since late April.
The next hard signal lands in September
The bullish case would weaken if the 10,000-plus cohort starts shrinking again or mid-sized wallets become net sellers over the next month.
The Senate is scheduled to return September 14, with a procedural vote on the CLARITY Act expected around September 15. That will provide a useful test of whether institutional and large-holder demand can continue absorbing supply, or whether the current accumulation trend is losing momentum.
Are larger holders gaining influence over Bitcoin’s supply?
Firstly, the address 14FEEMRhaUwMbhf2rA1cFXmS1Zuk9nc9eq has received 10,306.34 BTC in a single transaction on June 2 and has performed no outgoing transactions since then. Therefore, this wallet looks like it was just newly funded with more than 10,000 BTC, but it is impossible to attribute an economic owner based on the information available to the public.
On the other hand, the wallet bc1q7uq3u829ahn22sdlpac0h0lurq3a9yfd3ew69f had a balance of 7,269 BTC until it received 3,998.9 BTC on July 17, thus crossing the 10,000 BTC line. This wallet has received another 628 BTC after that, and now it holds almost 11,900 BTC.
This difference is quite important because in the first case, it could mean either a new large holder, a custodian transfer or re-balancing of the existing position. In the second case, one can clearly see that an increase in 10,000 BTC wallets does not always mean the appearance of a new whale: the existing large holder can just return over the threshold.
Therefore, the Santiment metric on 10,000 BTC wallets is still quite relevant, but only the number of the wallets is not enough to estimate how many independent investors are moving there.
A closer look at addresses around Bitcoin’s 10,000-BTC threshold suggests the rise in the whale count is not simply a wave of new investors. At least one address appears to have crossed the threshold after rebuilding an existing large position, while another was funded with more than 10,000 BTC in a single transfer.
An examination of addresses around the threshold shows why. One wallet was newly funded with more than 10,000 BTC in a single transaction, while another crossed the threshold only after rebuilding a position it had previously reduced. Neither case, by itself, proves that a new independent investor entered the market.
Are Bitcoin whales preparing for the next market move?
The signal that survives the scrutiny is therefore broader: Bitcoin is increasingly concentrated among large holders. Whether that reflects aggressive accumulation by independent whales, institutional custody and consolidation, or a combination of the two remains less certain.
For investors, that distinction matters. A market in which independent whales are accumulating represents one kind of bullish conviction. A market in which custodians, funds and existing large holders are simply reshuffling increasingly concentrated holdings represents another.
The next important metric may therefore not be the number of Bitcoin wallets holding 10,000 BTC, but the amount of supply controlled by identifiable economic entities after multiple addresses are consolidated. That is where the blockchain’s apparent whale surge will either become a genuine accumulation story—or reveal itself as a story about institutionalization.

The smartest crypto minds already read our newsletter. Want in? Join them.

Dyna Robotics on Monday introduced a new robot foundation model DYNA-2, which it says was trained on over a million hours of human video and no robot data.
The US-based company believes this approach could help solve the cost issues that have limited general-purpose robots.
Why robot builders keep running out of training data
DYNA-2 aims to solve a common problem in building general-purpose robots, which is getting enough data. Most robots learn through teleoperation, where a person guides the machine through tasks for hours.
The issue here is that the method is slow, costly, and hard to scale, which limits how advanced these systems can get.
Today we are introducing Dyna-2, a world-action model pre-trained on one million hours of human video. At this scale, for the first time, we discovered several new scaling laws:
• world-action models exhibit scaling law on human data across four orders of magnitude, from 1000… pic.twitter.com/wZamR0axzS
— Dyna Robotics (@DynaRobotics) August 10, 2026
Dyna’s solution is to leave out the robot during training. DYNA-2, instead, was trained only on egocentric human video, which is footage from a person’s point of view as they interact with objects.
The company says this adds up to about 170 years of continuous experience. Co-founder Jason Ma explained, “Action data is scarce, but video is everywhere,” arguing that physical intuition “can be learned directly from human video” instead of using millions of hours on a robot arm.
On high-precision manufacturing tasks, the success rates rose from 20% to between 80% and 90%. Dyna credits this improvement to the scale of pre-training.
Across 15 benchmark tasks, models trained on more human video consistently outperformed those trained on less human video.
What Dyna’s World-Action Model does differently
Dyna stands out because of its unique architecture. Instead of using a vision-language model, DYNA-2 is a World-Action Model based on video generation.
During pre-training, it predicts both the next video frame and the next action at the same time.
Dyna says this dual goal gives the model an internal sense of contact physics and spatial reasoning. The company hopes that knowledge from human videos will transfer to different robots, like stationary arms, humanoid prototypes, and dexterous hands, even though these devices were not used in pre-training.
Adapting to a new platform then takes just hours of fine-tuning instead of weeks of collecting new data. For example, Dyna reports that only 13 minutes of data taught robotic hands to twist off a bottle cap.
Dyna to extend approach to 10 million hours
Dyna already has robots in real-world settings, which sets it apart from other robotics firms.
DYNA-1 robots are being used in hotels, restaurants, and laundromats, including gyms, according to reports.
The company was founded by Lindon Gao, York Yang, and Jason Ma, who previously worked at DeepMind.
Dyna’s stated plan is to push the same approach to 10 million hours of video, which it frames as a collection problem rather than a fleet-building one.
If you're reading this, you’re already ahead. Stay there with our newsletter.

Trump Media & Technology Group (Nasdaq: DJT) told investors it lost $238 million in the second quarter, even as its Bitcoin stash grew large enough to overtake Elon Musk’s Tesla.
The $238 million loss is more than ten times the company’s reported loss in the same period last year. The main cause of the losses is falling crypto prices.
DJT overtakes Tesla
Bitcointreasuries.net now lists DJT at roughly 12,062 BTC, worth about $774.6 million as of August 11, just above Tesla’s holdings of 11,509 BTC, which have remained flat for years.
Top 16 publicly traded firms with Bitcoin holdings. Source: BitcoinTreasuries.net
Trump Media now occupies the 12th spot among public-company bitcoin holders.
In its Form 10-Q, Trump Media reported about 14,139 BTC as of July 31. That number includes every Bitcoin the company has promised to someone else, not just the ones it fully controls.
Bitcointreasuries.net removed 2,077 BTC from the total because it was posted as collateral for a covered-options strategy.
The filing says the counterparty “can rehypothecate at their sole discretion” those coins and any Bitcoin premiums it receives. Coins backing the firm’s $1 billion convertible notes stay in the tally, since that collateral remains recognized as Trump Media’s own.
The $238 million loss is more than ten times the shortfall Trump Media booked in the same stretch of 2025. The losses are reportedly tied to a slide in crypto prices, which also caused a $406 million first-quarter loss earlier in 2026 when Bitcoin fell below $70,000 after trading above $126,000 the prior October.
Accounting rules force the company to mark those holdings to market, whether or not it sells.
However, Trump Media posted $1.7 million in revenue for the quarter, which it said climbed 89% from a year earlier. The company closed the period with total assets of about $2 billion, with roughly $1.9 billion of it in financial assets, including cash, short-term investments, and digital currencies of about $1.2 billion in Bitcoin and Bitcoin-related holdings. It also carries $1 billion in convertible notes that mature in 2028.
Is Truth API contributing to Trump Media’s revenue?
Interim chief executive Kevin McGurn told the earnings call that Truth API charges $60,000 to $100,000 a month to push posts from Truth Social’s biggest accounts to institutional traders faster than the public sees them.
More than 10 customers have signed up for the service since it launched at the start of August, with most of them being high-frequency trading firms.
Cryptopolitan reported that the product drew legal and ethical questions, because Trump’s family remains the majority shareholder in a company that stands to profit when traders front-run the president’s own market-moving statements.
Kathleen Clark, a government-ethics expert at Washington University School of Law, said the arrangement amounts to selling privileged access to information about the President’s actions.
McGurn has rejected that framing, arguing that licensed real-time public data through commercial APIs is a well-established practice across technology and financial information firms.
McGurn also clarified that after a year of exploring the online betting and crypto markets, the company will refocus on social media, describing the shift as a “disciplined choice.”
Trump Media recently scrapped a plan with Crypto.com to add prediction-market features to Truth Social. However, the company is still pursuing its merger with the nuclear-fusion company TAE Technologies, which McGurn stated that he expects to close by year end.
The smartest crypto minds already read our newsletter. Want in? Join them.

Ravencoin, a proof-of-work chain worth around $57 million, warns about an attack exploiting a critical consensus flaw in which the compromised nodes will be able to validate blocks that shouldn’t be accepted. However, the problem with the attack isn’t limited to Ravencoin. Chain reorganization in a coin that’s worth relatively little could result in freezing funds already processed by exchanges and cross-chain bridges.
This means that the flaw is critical not just to the particular coin but to the whole cryptocurrency market. The transactions included in the blocks could be completely erased from the official chain when the old validated blocks are removed and replaced by the new ones. On Monday, the RVN price reached $0.003507, down by 1.2% compared to the previous day. Daily volume grew by 219% and reached $4.57 million.
What Ravencoin actually disclosed
On August 10, the Ravencoin team shared a notice on X asserting that a vulnerability has been “demonstrated and exploited.” As a result, invalid blocks may pass through the unpatched nodes. The first bad block recognized by the team is at block height 4,487,776, which is identified as having been timestamped at 15:44:01 UTC on August 7.
The period of three days from the time the first recognized invalid block occurred until the time it was made known is important because it is possible that transactions confirmed on a blockchain that was subsequently rejected by the new nodes will have to be undone.
Ravencoin is a Bitcoin fork released in January 2018 for the purpose of transferring assets on-chain. The developers of this project applied the KAWPOW proof-of-work algorithm with the aim of enhancing mining resistance against ASIC concentration.
The network has experienced a major consensus failure in the past. In 2020, a bug was exploited by attackers who were able to generate RVN beyond the acceptable block subsidy. The lead developer of Ravencoin, Tron Black, said later:
“The vulnerability does not allow the stealing of RVN or assets that you own and control.”
According to Black, these extra coins have been transferred to an exchange where they have been combined with legitimate RVN. Thus, a mere rollback is no longer possible. The total excess issuance was approximately 301.8 million RVN.
Why small-cap chains keep getting hit
Networks with low market capitalization and mining are believed to have a poor economy. An article in Complex & Intelligent Systems published in February 2026 discovered that blockchains that are new and have market capitalization below $100 million may have attack costs ranging between $50,000 and $1 million, while established blockchains would need billions of dollars.
According to the article, after its review of documented 51% attacks between 2018 and 2024, it was discovered that 85% of the successful 51% attacks were carried out on blockchains in the new phase, where 80% of the attacks were successful.
Moreover, they researched the characteristics of checkpointing and Byzantine Fault Tolerance (BFT) in terms of defense. The conclusion drawn by them was that systems which were developed based on BFT concepts had some certainty of finality and were more secure than the traditional proof-of-work concept.
Ethereum Classic demonstrated the possible ramifications of such an attack. It was attacked four times successfully in 51% of reorganizations – once in January 2019 and three times in 2020.
The Litecoin warning shot
Litecoin offers a recent example of how reorg damage can spread beyond the affected chain. In an April 28, 2026, postmortem, developer David Burkett said a Mimblewimble Extension Block bug produced a 13-block invalid chain during a second exploitation attempt in April. Upgraded miners eventually coordinated on the valid chain and reorged the invalid blocks out.
But some third-party cross-chain services had already processed transactions. Litecoin’s official postmortem said NEAR Intents processed a swap of 11,000 LTC for 7.78814476 BTC, leaving it with a loss after those LTC disappeared from the valid chain. THORChain suffered a separate loss after processing 10 LTC for 0.00719957 BTC.
This is the risk Ravencoin is currently facing. The breach calls into question the integrity of the chain and ultimately could have damaging effects on exchanges, bridge services, and swap services that may confirm transactions prior to the final version of the chain being verified. Until operators of the affected platforms confirm that they are using the patched software, RVN transactions may be subject to the risk of reorganization over which individual holders have no control.
Don’t just read crypto news. Understand it. Subscribe to our newsletter. It's free.

Riot Platforms shares jumped over 20% before U.S. markets opened on Tuesday, following reports that Anthropic is the buyer behind a $9.1 billion, 20-year deal to lease AI computing capacity from the Bitcoin miner’s Texas campus.
When Riot Platforms announced the deal on Monday, it did not reveal the other party, calling the tenant only “one of the world’s leading frontier AI labs.” Hours later, Bloomberg reported that the customer was Anthropic, citing people familiar with the matter.
After the report, retail trading activity picked up. RIOT became the most-discussed ticker on Stocktwits, and sentiment changed from ‘bullish’ to ‘extremely bullish.’ The stock gained over 26% overnight. And even before Monday’s jump, the stock was already up 53% for the year.
RIOT YTD price chart. Source: Yahoo Finance
Riot leases 191 megawatts at Rockdale through 2028
The contract includes 191 megawatts of capacity at Riot’s Rockdale, Texas site and lasts for 20 years. Two possible five-year extensions could raise the total value to $16.1 billion.
Riot Platforms expects the base term to generate between $7.3 billion and $8.2 billion in net operating income.
The new capacity will be added in phases. Up to 96 megawatts will go live in December 2027, with full deployment expected by June 2028. Riot said a $573 million interim loan from Morgan Stanley will cover early development costs until a long-term credit backstop is in place.
Anthropic now becomes the second tenant at Rockdale. In January, Riot had also signed a deal with AMD, which saw the shares rise 13% on the news, as Cryptopolitan reported.
With both deals, Riot’s contracted AI capacity at the site now totals 241 megawatts.
In the company’s earnings statement, CEO Jason Les said they have “now executed leases totaling 241 megawatts of capacity, representing approximately $9.8 billion of long-term, contracted revenue with two of the most important companies in the AI ecosystem.”
Bitcoin mining revenue dropped as Riot shifted its focus.
The news comes at the same time as Riot’s mixed second-quarter results. Revenue increased by 14% to $174.2 million, but Riot reported a net loss of $237.2 million, compared to a $219.4 million profit a year ago, according to Yahoo Finance.
Data center revenue was $23.2 million, with $4.9 million from operating leases and $18.3 million from tenant fit-out services.
Revenue from mining fell to $113.7 million because lower prices and more competition offset higher production, alongside falling BTC prices.
Riot has been selling its monthly Bitcoin output and reducing its treasury to help pay for data center expansion. Its holdings fell from 15,680 BTC to 11,380 BTC during Q2, a decrease of 4,300 coins.
Don’t just read crypto news. Understand it. Subscribe to our newsletter. It's free.