Akash's migration to Solana may enhance Solana's ecosystem appeal, potentially influencing developer interest and investor confidence.

The post Akash Network to migrate from Cosmos to Solana with 98% approval appeared first on Crypto Briefing.

Anthropic's improved revenue efficiency may boost investor confidence, but ongoing unprofitability could affect its market valuation and perception.

The post Anthropic boosts revenue efficiency to $2.08 per $1 spent on AI compute appeared first on Crypto Briefing.

Akash Network's shift to Solana could boost AKT's liquidity and integration within a vibrant ecosystem, impacting DePIN sector dynamics.

The post Akash Network proposal to leave Cosmos for Solana shows ~98% support, AKT to become SPL token appeared first on Crypto Briefing.

Strive's aggressive Bitcoin acquisition strategy highlights the intensifying competition among corporate treasuries, impacting market dynamics and investment strategies.

The post Strive is 1,001 Bitcoin away from surpassing Metaplanet as second-largest Bitcoin treasury appeared first on Crypto Briefing.

The incident highlights the need for stricter monitoring and network restrictions in AI deployments to prevent unintended data transfers.

The post Cohere says Alibaba’s Qwen model tried to send data to a Beijing server during a security test appeared first on Crypto Briefing.

Bitcoin Magazine

Blink Wallet Adds Inactivity Fee to Custodial Accounts, Accelerating Migration to Self Custody

Blink Wallet announced a new $1 a month inactivity fee, aimed at custodial accounts that have not logged in to the app in over 12 months. 

The decision was put into motion in early September via a Terms of Service change, with push notifications sent out to inactive users, and the first charge made by mid-October. Fees charged will be refunded if users log back in within 12 months, according to a blog post published by the company. 

The fee, which only affects custodial accounts, serves to relieve the burden of inactive accounts on the startup. According to the announcement, these inactive accounts have ongoing costs such as “ledger and infrastructure, retention of verification records, compliance monitoring and reporting, and support claims.” These costs are in turn borne by active user accounts via normal in-app transaction fees. 

A source familiar with the matter told Bitcoin Magazine that around a hundred thousand such accounts fit the inactive category at the time of the first fee charge, with an aggregate balance of over $800,000. The first fee pulled in about $45,000, mostly from accounts that had dust amounts of bitcoin inactive for over a year. None of these figures could be independently verified. 

A Bit of History

Blink Wallet may be one of the most popular Bitcoin wallets in the world, favored by advocates of Bitcoin Circular Economies across Latin America and Africa. Designed for an audience not already familiar with Bitcoin, the wallet first came to prominence as the ‘Bitcoin Beach Wallet’. It grew out of El Salvador’s initial wave of Bitcoin adoption, choosing design principles similar to Wallet of Satoshi, which had seen viral success in earlier years. The wallet was initially custodial and oriented around fast Lightning payments, with the innovative option of hedging value in USD via the Stable Sats feature.

Historically, custodial wallets have seen accelerated early growth compared to their non-custodial counterparts. It is arguably due to the friction introduced by non-custodial setup steps like the 12-word backup, but also by the lack of key material under the company’s control, which enabled features like the first generation of Stable Sats in the first place.

Because of the smooth interface and user experience Blink offered, it quickly became a favorite for evangelists to recommend. One particular evangelist allegedly got everyone he met to download the app and gave them about $1 worth of bitcoin. Some of those new users might have stuck around. But events like today’s Blink announcement put into question the effectiveness of such evangelism strategies, given that they may just be turned into dormant dust accounts. 

Blink was eventually spun off from Galoy Inc its founding firm, to be run by Lukas Duczko and a new board of directors. The announcement published in October 2024 stated that Duczko now had a controlling stake in the company, while Galoy continued to be a service provider and minority stakeholder. The Bitcoin Beach Fund was also established at the time, seeded with Blink shares as part of the transaction. The wallet once headquartered in El Salvador is now based in the free economic zone of Roatán, Honduras, though it maintains corporate entities in El Salvador and the United States.

Headwinds

Despite its success gaining international adoption among Bitcoin circular economies, evangelists and supporters of the Lightning Network, the company has also faced significant headwinds. And not just Blink but custodial wallets in general, like Wallet of Satoshi, Alby, and Blue Wallet, all phased out their custodial wallet offerings in recent years, pivoting to self custody infrastructure like Spark. The tide began to rise with the EU’s MiCA regulations, which sought licensing and KYC compliance standards from custodians that these wallets simply could not provide. 

On the U.S. side of the pond, the Samourai Wallet case and the broader war on crypto by the Biden administration pushed wallet operators outside of the country, with notable examples like Phoenix Wallet, Wallet of Satoshi, Wasabi Wallet, and Blink, limiting or altogether exiting American app stores. The legal uncertainty took place under the shadow of the Samourai Wallet and Tornado Cash cases, which the industry read as threats to user privacy and certainly to self custody wallets, let alone custodial ones. 

Fast forward a couple of years and Google Play decided to enforce EU and American laws by limiting Play Store distribution of wallets or exchange apps that do not hold licenses in corresponding jurisdictions. The viral nature and stunningly easy-to-use apps that were possible thanks to custodial Lightning had, in a matter of 2 years, become a liability. It was time to pivot to self custody. 

Nobody Said it was Easy

By 2025, Blink’s counterparts already offered self custody options; WOS was beginning to push their Spark-powered wallet, with 2027 as the deadline for custody wallet users to migrate. Alby had shipped a feature to connect self-hosted Lightning nodes to its hub and shut down its custodial Lightning option. Blue wallet had long ago shut down its custodial offer and had since doubled down on becoming an onchain wallet for new and power users alike.

Unlike its competitors, however, Blink’s biggest differentiator is the Stable Sats feature, which likely required a significant redesign to fit the limitations of self custody. It wasn’t until June 2026 that Blink launched its Spark supported self custodial wallet, a big move and one that its user base is still adapting to. 

To make things worse, on September 19th, hackers managed to withdraw 6.61 bitcoins from custodial wallet users via compromised admin tools. The company’s backers quickly announced they would make the affected users whole, and the bleeding was stopped, but the blow nevertheless must have shaken the company. 

Remarkably, despite some of the most difficult years for the Bitcoin wallet industry, Blink wallet has so far not just survived but stayed in the game. It has retained the slick design that made it so popular, while delivering advanced features that users in developing nations need, such as dollar-denominated value at the speed of lightning, meanwhile pivoting to self custody. The inactivity fee is but an inevitable step further down that transition. Perhaps having teams like Blink’s working on self custody will yield a new crop of sovereign technologies, its fruits badly needed by a sector of the industry battered by regulators and now AI cyber threats. 

This post Blink Wallet Adds Inactivity Fee to Custodial Accounts, Accelerating Migration to Self Custody first appeared on Bitcoin Magazine and is written by Juan Galt.

▶

Bitcoin Magazine

Coinbase Institutional Co-CEO: Bitcoin’s “Second Wave” of Institutional Capital has Arrived

Seven of the largest banks in the world are now running on Coinbase infrastructure, and Brett Tejpaul, Co-CEO of Coinbase Institutional, says they’re having a “Kodak moment.” He joins Grace Remington and Sean Hagan to explain why the driver has shifted from Bitcoin ETFs to stablecoin payments now that money can move from a Citi account to USDC, across the blockchain in a second, and into a corporate account on the other side. He also gives a candid read on Bitcoin: the macro setup is great, the performance has been disappointing, and institutions aren’t yet treating it as anything but a tech trade.

Chapters:
0:00 Seven of the World’s Largest Banks Are Using Coinbase Infrastructure
1:19 From ETFs to Payments: Citi, USDC and Instant Bank-to-Blockchain Transfers
2:44 $400 Million an Hour in Stablecoins and the Banks’ “Kodak Moment”
4:38 Stablecoins Are a “Speck of Dust”: Payment Providers First, Then Banks
6:10 Agentic Commerce Arriving at the Same Time as the Big Players
6:49 Is Bitcoin Still a Leveraged Tech Trade? “Disappointing” Performance, Great Setup
8:44 Connecting U.S. Funds to Deribit’s 85% Options Market Share
11:06 The Second Wave: Net-New Entrants and Coinbase’s “Amazon to AWS” Pivot
12:57 Activating the Whole U.S. Through Coinbase Infrastructure
13:56 Quantum, 90% ETF Custody Share and “Safe Forever”

DISCLAIMER: The views and opinions expressed in this show are those of the participants and do not necessarily reflect the official policy or position of BTC Inc., Bitcoin Magazine, or any affiliated entities. This content is provided for informational and educational purposes only and should not be construed as investment, legal, tax, or accounting advice. Nothing contained in this show constitutes a solicitation, recommendation, endorsement, or offer to buy or sell any securities or financial instruments. Viewers should consult their own advisors before making financial or business decisions.

This post Coinbase Institutional Co-CEO: Bitcoin’s “Second Wave” of Institutional Capital has Arrived first appeared on Bitcoin Magazine and is written by Patrick Green.

Bitcoin Magazine

Bitcoin Price Hovers Near $80,000 as Traders Await Fed Decision

Bitcoin price traded near $80,000 on Monday, after falling about 2% earlier in the session. The price has dipped below that level several times in recent days and recovered each time.

Selling pressure has been building. Spot Bitcoin ETFs recorded about $635 million in outflows on October 22. A brief drop below $80,000 last week also triggered roughly $800 million in liquidations of long positions, or bets that prices would rise.

The next major test to look out for is the Federal Reserve’s two-day meeting, which ends October 28. The Fed raised its target range to 3.75% to 4% in September, and its minutes suggested another increase could come by year-end. 

Most recent market odds favor the Fed holding rates steady, though those odds have shifted. Traders are watching Fed Chair Kevin Warsh’s remarks for signs of a December hike.

Higher interest rates typically pull money away from risky assets, including crypto. Broadly speaking, a hold could keep bitcoin price between roughly $80,000 and $88,000. A rate hike could push it toward the lower end of that range.

Bitcoin price backdrop

Geopolitical tensions are also weighing on markets. Oil rose above $100 a barrel in early September as strikes between the U.S. and Iran escalated. Higher energy prices feed into inflation, which affects the Fed’s decisions. 

In tech, Anthropic is reportedly preparing an initial public offering that could value the AI company near $965 billion. The timing has shifted toward the weeks before the November midterm elections, but no date has been confirmed. 

U.S. stocks surged a bit Monday as investors geared up for a busy week of Big Tech earnings and the Federal Reserve’s interest-rate decision. The Nasdaq rose 0.8%, the S&P 500 gained 0.6%, and the Dow climbed 0.3%. 

Alphabet and Microsoft report Wednesday, followed by Amazon Thursday, while markets largely expect the Fed to hold rates steady. 

AI-related stocks rallied following Anthropic’s IPO filing. Tesla jumped 6%, while Meta plans to raise up to $95 billion for AI investments. 

As of writing, bitcoin price trades at $80,215, down nearly $4,000 or 4.75%, on the month. Over the past month, prices have ranged from a high of $86,806.95 to a low of $79,397.76. Bitcoin price peaked above $86,000 in early October, then slid toward $80,000 in the second week of the month. 

A brief rally around October 21 pushed it back near $84,000, but those gains faded within days. Trading volume over the past 24 hours reached about $18.12 billion. 

Bitcoin price remains far below its all-time high of $126,277.05.

This post Bitcoin Price Hovers Near $80,000 as Traders Await Fed Decision first appeared on Bitcoin Magazine and is written by Micah Zimmerman.

Bitcoin Magazine

Strategy (MSTR) Pauses Bitcoin Buying and Spends $205 Million on Preferred Stock

Strategy reported no bitcoin purchases or sales during the week of October 19 through 25. The company continues to hold 848,517 BTC, acquired for $64.02 billion at an average price of $75,446 per bitcoin. 

Its cash went toward repurchasing $204.9 million of STRC preferred stock, and the firm kept a $4.83 billion cash reserve, according to a company release. 

Michael Saylor’s company built its treasury into the largest corporate bitcoin holdings in the world through stock sales and convertible debt, using the proceeds to buy coins. That pattern has changed somewhat this year. The company has paused purchases several times and directed money toward buybacks of its preferred shares.

The company has had stints of pauses, from a week when the company halted its bitcoin buys to a stretch in which it sold bitcoin after five weeks without buying.

In September, the company repurchased $176 million of STRC in one week and expanded its digital credit securities repurchase program from $1 billion to $2 billion.

Strategy’s preferred dividend

STRC is a preferred stock with a variable dividend, designed to trade near its $100 face value. When the shares fall below that level, as they have during recent market swings, buying them back shrinks the preferred claim. Retiring $100 of stated value for $86.52 cuts the claim by a larger amount than the cash spent, which lifts net bitcoin per share.

Bitcoin traded near $80,000 on Monday, October 26, after slipping below that level in early Asian hours. A weekend selloff pushed the price to the low $79,000’s, a level not seen in over a month.

Strategy’s average cost of $75,446 sits below that price, so the company’s holdings show a paper gain of around $3.9 billion. The $4.83 billion reserve covers preferred dividends and debt interest, according to company reporting. 

Strategy’s buyback capacity depends on preferred stock trading below par and on cash from common stock sales, a funding mix that has shifted toward stock sales this year. 

The company has offered no timetable for resuming bitcoin buys.

This post Strategy (MSTR) Pauses Bitcoin Buying and Spends $205 Million on Preferred Stock first appeared on Bitcoin Magazine and is written by Micah Zimmerman.

Bitcoin Magazine

Bitcoin’s Bear Market Is Over, Galaxy Says

Galaxy Research believes bitcoin’s bear market ended in June and that the asset has entered an accumulation phase, according to a research alert from the firm.  “We believe now is an attractive time to buy and hold bitcoin,” the report said.

Bitcoin traded near $80,000 at the time of the note, about 35% below its October 2025 high and 39% above the June 30 low. The price would need to rise 53% to retake the previous record. The S&P 500, by comparison, sits within 1% of a record set on Oct. 6.

The report lays out four arguments: technical, positioning, fundamental and sentiment.

Bitcoin technicals

Galaxy uses two weekly moving averages to track bitcoin’s bear markets. The 200-week average has acted as a floor in each one, and the 50-week average has acted as a ceiling, according to the firm.

Bitcoin lost the 50-week average in November 2025 and stayed below it for 44 weeks. The June low tested the 200-week average and held, with the worst weekly close coming 4.7% beneath it. Price closed back above the 50-week average in September, and the week ended Oct. 18 was the fifth consecutive weekly close above that level.

Galaxy counted 13 such reclaims since 2011 and said the outcome depended on how long price had spent below the line. The four reclaims that ended spells of 20 weeks or more were higher a year later, by between 55% and 1,497%, and none closed more than 8.4% below the reclaim level in the following 12 months. Of the nine that ended shorter spells, seven fell more than 25% within a year.

The firm said the halving cycle supports the same conclusion. Prior cycles bottomed on days 777, 889 and 912 after their halvings. Oct. 22 was day 916 of the current cycle. Galaxy said the gains that followed those lows (87x, 19x and 7.6x) have shrunk each cycle and that it does not forecast a repeat.

The current drawdown of 53% is the shallowest in bitcoin’s history, the report said, and the first in which price did not fall below the network’s realized cost basis.

Positioning

The report identified basis traders and macro funds as the main sellers since the top. Asset managers’ net long position in CME bitcoin futures fell 57% from the week of the all-time high, according to Commodity Futures Trading Commission data cited by Galaxy. 

Leveraged funds’ net short position, which represents the futures side of the cash-and-carry trade against spot ETFs, shrank by a similar amount. The front-month basis averaged 7.9% annualized in the 30 days before the high and 4.8% in the 30 days to Oct. 21, which the firm said no longer covers the trade’s costs and risk.

Galaxy pointed to 13F filings to identify the sellers. Millennium Management, Brevan Howard, Goldman Sachs, Capula, Farallon, D. E. Shaw, Schonfeld, Symmetry and Tudor cut their spot ETF holdings by between 29% and 100% between the September 2025 and June 2026 quarter-ends, a combined 68,000 BTC. 

Because total holdings across all 2,200 filers fell by 48,000 BTC, the remaining institutions added in aggregate. Morgan Stanley’s wealth management arm, Mubadala and the Abu Dhabi Investment Council were among the buyers.

Spot ETF holdings dropped 11% from peak to trough while price fell 53%, and the funds have added 87,000 BTC since the June low, the report said.

Fundamentals and sentiment

Galaxy’s fundamental case centers on U.S. fiscal conditions. Federal debt reached $40.44 trillion on Oct. 20, fiscal 2026 ended with a $1.99 trillion deficit, and the 10-year Treasury yield hit 5.31% on Oct. 5, its highest since 2007. 

The firm argued that a government refinancing that debt at 5% while running 6% deficits faces pressure to inflate.

The report described AI-related borrowing, which Morgan Stanley expects to reach $570 billion this year, as a contributor to that pressure rather than a competitor for capital.

On sentiment, Galaxy compared the current environment to 2015–16 and 2018–19, when mainstream finance embraced “blockchain” and dismissed bitcoin. Both periods coincided with cycle lows. The firm said the 2026 version is “stablecoins and tokenization, not bitcoin,” citing bank pilots of tokenized deposits, the Senate’s failed cloture vote on the Clarity Act on Sept. 15, and Google searches for “bitcoin” at multiyear lows.

Risks

Galaxy set a tripwire for its view: a weekly close below the 50-week average, about $76,540, that is not regained within a month. It cited AI-enabled theft, including the $387.5 million Bitget exploit, while noting none of the incidents involved the Bitcoin protocol itself.

On quantum computing, the report said 7.1 million BTC sit in outputs with exposed public keys but that no cryptographically relevant quantum computer exists. It pointed to BIP-360’s merge into the BIP repository in February and Galaxy’s own $5 million quantum readiness commitment.

This post Bitcoin’s Bear Market Is Over, Galaxy Says first appeared on Bitcoin Magazine and is written by Micah Zimmerman.

An already-compromised signer could recover the key through many malicious ECDSA messages when the application ignored a critical stop condition.

Read the full story on CryptoSlate

Understand the crypto stories that matter and what to watch next. Get The Catalyst in your inbox

Ripple expanded its institutional reach Monday as Aviva added another tokenized fund share class to the XRP Ledger (XRPL). The British asset manager invested £50 million ($66 million) in a blockchain-based share class of its Sterling Liquidity Fund, bringing its combined investments across two XRP Ledger offerings to more than £86 million ($116 million). The…

Read the full story on CryptoSlate

Understand the crypto stories that matter and what to watch next. Get The Catalyst in your inbox

Bitcoin treasury firms Strategy and Strive made record purchases last week, but directed their capital toward different assets. While Michael Saylor's Strategy intensified its preferred-stock buybacks, Strive accelerated its Bitcoin accumulation, highlighting diverging priorities in how the companies deploy capital. Strategy's STRC purchases exceed $2 billion Strategy's Bitcoin buying streak ended after five consecutive weeks…

Read the full story on CryptoSlate

Understand the crypto stories that matter and what to watch next. Get The Catalyst in your inbox

FHLBNY’s October 23 update puts dealer financing capacity in focus ahead of October 30, while Bitcoin spillover remains conditional.

Read the full story on CryptoSlate

Understand the crypto stories that matter and what to watch next. Get The Catalyst in your inbox

Gateway gives cirBTC holders a shared balance, with advertised fast access after deposit finality and a separate seven-day fallback exit.

Read the full story on CryptoSlate

Understand the crypto stories that matter and what to watch next. Get The Catalyst in your inbox

Ethereum trades at around $2,308 on Monday evening and closes its fourth trading day in a row in the red. The price sits below the 200-day average of $2,329, and that is the real finding of the day: this line has carried every recovery of the current year.

Documented up to this point is the following: the price of $2,308, down 1.7 percent in 24 hours, the two moving averages from the chart further down, October's balance of minus 14.1 percent and the US Federal Reserve's rate decision on Wednesday. What the editorial team concludes from that for the coming days has a section of its own further down and is marked there as an assessment.

Where the price can point on a horizon of months is what our Ethereum forecast tracks continuously. This article takes today.

Where Ethereum stands on Monday evening

The day brought a narrow range. The high was $2,347 and the low $2,293, with 2.4 percent between the two. Ethereum's market capitalization comes to $282 billion, and trading turnover over the past 24 hours to $10.7 billion.

What decides, though, is not the single day but the run. Four daily closes in a row came in below the one before: $2,473 on October 22, then $2,386, $2,385 and $2,357, and today $2,306. Ethereum last traded that low on August 20, at $2,252 back then. The current month stands at minus 14.1 percent, counted from October 1 at $2,685. From the monthly high of $2,726 on October 5 it is minus 15.4 percent.

Bitcoin gave up less over the same period and stands at $79,776, down 0.9 percent in 24 hours. Ethereum is therefore losing faster than the market it was still leading in the summer.

What the 200-day line measures for Ethereum

The 200-day average is the mean of the daily closing prices of the past 200 days. It says nothing about what happens tomorrow. It says where the middle of the past nine months lies, and it shifts slowly, because each new day changes only a small part of the mean. That is exactly why traders watch it: it is a line many see at the same time and one that does not move with every daily swing.

For Ethereum that average sits at $2,329 today, the 50-day average at $2,457. The price is below both. That the short line runs above the long one also shows that the gap has opened only in recent weeks; in the summer the price was still well above them.

The chart also shows the year's span. The high was $4,159, the low $1,566 at the end of June. Measured from the high, Ethereum stands 44.5 percent lower, and 47 percent higher than the June low. Both figures belong together when placing today's level: the price is neither on the floor nor near its best mark.

Three reasons for the fourth day in the red

The first reason is the calendar. The US Federal Reserve meets on October 27 and 28, with the rate decision following on Wednesday evening. Ahead of such dates larger players take risk off, because the direction is open. That hits Ethereum harder than bitcoin, because the market there is thinner.

The second reason is the relative weakness against bitcoin. One ether currently costs 0.0289 bitcoin. As long as that ratio falls, capital is flowing away from Ethereum within the crypto market, and that works regardless of what the dollar price does.

The third reason is the line itself. When a closely watched level gives way, automatic sell orders sitting beneath it trigger. That amplifies a move which began for other reasons, and it explains why the slide of the past four days managed without a single large piece of news.

Buy or sell in this situation and you pay the spread between bid and ask along with it. On thin days it is wider than usual. A look at the crypto exchange comparison is therefore worth it right now, because fee structures differ markedly at small amounts.

The Fed decision on Wednesday and what it means for Ethereum

The meeting of the US Federal Reserve's open market committee runs on October 27 and 28, as set out in the Federal Reserve's meeting calendar. The connection to crypto runs through the interest rate: if it falls, money becomes cheaper, and assets without running income become relatively more attractive. If it stays higher than expected, the sum works out the other way.

For investors in Germany a second point matters here. Ethereum is traded in dollars, while your portfolio counts in euros. If the dollar falls after a rate decision, part of the dollar gain melts away in the conversion, and the other way round. This currency effect is no footnote: on a price move of a few percent it can account for half the result.

Ethereum measured against bitcoin

The ratio of ether to bitcoin is the figure that shows the direction within the crypto market. At 0.0289 bitcoin per ether it sits close to the lower edge of recent weeks. If you hold Ethereum because you are convinced by the blockchain and its applications, you take that number as a cost item: choosing Ethereum over bitcoin has cost money this month.

That is a different view from the one on the dollar price, and it leads to a different question. Counting in dollars, you ask whether the market as a whole turns. Counting in the ratio, you ask whether Ethereum converts its lead in applications back into price. Both questions share the same date on Wednesday, but not the same answer.

The one-year deadline on October 27 and loss offsetting

Here a date comes into play that has nothing to do with the chart and can still matter more for German investors. On October 27, 2025 Ethereum stood at $4,159, the high of the past twelve months. Anyone who bought on that day reaches the one-year deadline under Section 23 of the Income Tax Act tomorrow, while sitting on a loss of 44.5 percent.

On a gain the sum is simple: after a year of holding, the gain from a private disposal is tax-free. On a loss the logic turns around. Sell within the year and the loss counts for tax, and you can offset it against gains from other private disposals in the same year. Sell after the year has elapsed and the loss is irrelevant for tax, because a transaction outside the period is not taxable at all.

In concrete terms: if you hold Ethereum at a loss, stand shortly before the one-year deadline and have realized gains from other crypto sales this year, you have room to act before the deadline and none after it. If you are in profit, by contrast, selling before the deadline costs you the tax exemption. Which purchase date belongs to which position is shown by a clean record of your acquisitions; the tools for that are in the comparison of crypto tax tools.

Whether the one-year deadline stays in this form is politically open. Our page on the crypto holding period and its reform tracks the state of the debate and the deadlines that currently apply. For this year the one-year period applies unchanged, and the threshold for private disposal gains is 1,000 euros.

Why the purchase date decides and not the price level

The tax question hangs on the purchase date. Two investors with the same loss can therefore reach different outcomes if one bought in September 2025 and the other in February 2026. That is why a losing position is not one uniform case but as many cases as there are acquisitions.

The frame below and above

Below, the next justified level is the August 20 reading at $2,252. That is where the price sat the last time Ethereum traded this low, and the late-summer recovery began from there. Should that area not hold, the next orientation lies considerably lower, because between $2,252 and the June low of $1,566 the price has spent barely any time this year.

Above, two values matter, and both are in the chart. The 200-day average at $2,329 is only 1 percent away; reclaiming it would be a first sign, no more. It becomes solid only at the 50-day average of $2,457, because that is the area the past week's slide started from.

What has changed since Sunday's Ethereum article

On Sunday the gas fee was at the centre here. The analysis of 9.7 ETH burned in 24 hours dealt with how little ether the blockchain takes out of circulation when usage is thin. The price stood at $2,357 at the daily close of October 25.

Two things have come on top since. First, the price has given up a further 2.1 percent and thereby undercut the 200-day line, which still held on Sunday. Second, the Fed date has turned from next week's news into this week's news. The gas fee itself has not turned; Sunday's thin usage remains the backdrop against which the price is falling.

The editorial team's assessment

Our assessment: the break of the 200-day line is a weak signal so far, not a strong one. In favour of that reading is the gap of only 1 percent, the narrow daily range of 2.4 percent and bitcoin's loss of 0.9 percent, which shows that no market-wide sell-off is under way here. Against it stand the run of four days in the red and the falling ratio of 0.0289 bitcoin per ether, both signs that capital is not coming back just now.

This is decided on Wednesday, and not by the chart. A rate decision that calms the market turns the break into a four-day episode. If the rate stays higher than expected, the area around $2,252 gets tested this week. Predicting a direction without knowing the decision would be guesswork, and guessed numbers do not appear in this article.

Ethereum: the trend only turns above $2,457

Three steps make sense this week, and they do not depend on where the price runs.

1. Check in your records which Ethereum acquisitions are at a loss and when their one-year deadline expires. A loss can be used for tax only on acquisitions inside the period; the tools for that are in the comparison of crypto tax tools.
2. Compare your platform's fees and trading spread before you trade again ahead of Wednesday. On thin days the spread costs more than the order fee, and the differences are in the crypto exchange comparison.
3. If you want to hold Ethereum for longer, check where the holdings sit. For amounts you do not want to touch beyond Wednesday, self-custody is the calmer solution; the devices are in the hardware wallet comparison.

(As of October 26, 2026. This article is not investment advice. Prices and fee structures change; check the terms with the provider before you buy.)

A hardware wallet protects your keys reliably. It does not protect your home address. The short answer to getting both at once: have the device delivered to a parcel locker, a post office branch or another pickup point, or buy it over the counter in a shop. Your name then sits in the retailer's order database next to a locker number instead of next to your front door.

October showed that this difference is no quirk of the overcautious. Within a few months several providers in the custody business have conceded that customer data leaked. Not once did it involve keys or balances. Every time it involved names, phone numbers and delivery addresses of people for whom it is now on record that they own crypto assets.

Why the keys stay safe and the address leaks anyway

A hardware wallet is a small device that generates your private key and never hands it out again. The recovery phrase, that sequence of usually twelve or twenty-four words from which all keys can be recomputed, is created on the device and does not leave it in normal use. When you approve a transfer, the key stays on the device; only the finished, signed transaction travels to the computer. The retailer you bought the device from sees none of this, because technically it is not involved at all.

That leaves one reassuring fact and one uncomfortable one. The reassuring one: a break-in at a manufacturer's web shop cannot touch your balances. If your bitcoin has sat on such a device for years, a shop mishap means nothing to move and nothing to set up again. The uncomfortable one: the web shop itself is an ordinary online retailer with an ordinary customer database. It holds first name, surname, street, house number, city, country, often the phone number, the email address and the list of items ordered along with prices.

That combination is the problem. A stolen address list from a furniture store reveals that somebody bought a sofa. A stolen address list from a wallet maker's shop reveals that somebody keeps crypto assets at home, how expensive the model was and therefore roughly how serious they are about it. The data set is a shortlist an attacker would otherwise have to assemble laboriously.

Keystone and Trezor: where in the order chain the data sat

Keystone says it noticed the incident on October 20 and confirmed on October 23, after checking with its shop provider Shopify, that a third party had obtained customers' shipping information. The company made the case public on October 24, together with an apology. According to the firm, names, contact details, delivery addresses and order details down to items and prices are affected. Payment card data, bank details, recovery phrases, private keys and wallet passwords did not leak, it said. Keystone has not said to this day how many customers are affected. The particulars of this case are in our report on the confirmed data breach at Keystone.

At Trezor the leak sat one layer further out. The manufacturer announced on August 13 that data of 13,689 customers had been exposed after its logistics provider ShipMonk reported unauthorised access three days earlier. For 11,742 of those affected this covered name, delivery address, phone number and email address, and for a further 1,947 at least name, city and email address. It hit orders shipped between May 10 and August 8, among others to the United States, the United Kingdom, Sweden, Italy and Portugal. The more awkward half came in early September: the stolen files also contained old order data from an earlier collaboration in the years 2019 to 2021, around 67,000 further records in the United States alone. According to Trezor, that legacy data should have been deleted long ago under the contract. What those affected can do about it is written up in our piece on the Trezor leak.

Both times the finding is the same: the devices and the manufacturers' own systems remained untouched. What leaked was the chain that carries an order from the input form to the front door.

Shop platform, warehouse provider, analytics tool: who stands between maker and letterbox

Order a device online and you have a business relationship with exactly one company. Your data, though, passes through several. First comes the shop platform the store runs on at all; Keystone names Shopify here, and Shopify was already at the centre of the largest case of this kind to date back in 2020. Names, email addresses, postal addresses and phone numbers of around 272,000 Ledger customers were published then, a figure the company later corrected to about 292,000 because Shopify reported further victims. The data surfaced in a hacker forum in December 2020, together with more than a million email addresses. Neither keys nor balances were affected that time either, and that was of little help to those concerned: at least one buyer later reported moving house after threats.

Then there is the warehouse and shipping provider that packs the goods and hands them over. It needs the full address, or it cannot deliver anything. In ShipMonk's case, a third layer came into it according to the notification letters seen by the trade outlet BleepingComputer: the attackers exploited a weakness in the analytics tool Metabase, software the provider uses for its own figures. The buyer of a hardware wallet has never heard of that software, and yet their data lay within its reach.

The pattern is therefore easy to describe. The longer the chain between order and letterbox, the more places hold the same address, and the more contracts stand between you and the company that eventually loses it. No manufacturer can guarantee that every link in that chain works cleanly. You can, however, make sure the address running through the chain is not the one where you sleep.

Four ways to order, and what each one tells the retailer

All four routes cost the same for the device itself. What differs is only the data set left behind at the end. The first runs via a pickup station, the second via a branch of the delivery firm, the third via buying in a shop, and the fourth does without a physical device. All four are readily available in Germany, and none asks you to misrepresent yourself to the retailer or give false details. That would be neither necessary nor wise, because warranty and returns depend on correct data.

A parcel locker accepts the package without your home address

The most direct route is a pickup station. At DHL it is called the Packstation, and other carriers run pickup points under other names on the same principle: you get a customer number, enter the number of the station as the delivery destination and collect the parcel when it suits you. The retailer's database then holds your name, a post number and the address of a machine standing in a supermarket car park. An attacker who gets hold of that list knows you bought a hardware wallet, but not where you live.

For recipients there is no surcharge, and registration runs through the carrier's customer account. Two limitations come with it. First, your real address still sits with the carrier itself, because the customer account knows it; you are moving it from many places to one. Second, pickup stations have compartment sizes and storage deadlines, which with a device the size of a cigarette packet is rarely a problem.

Branch delivery and the post office: the route without an account at the carrier

If you would rather not register anywhere, you can have the parcel sent to a branch. The delivery address is then the branch's address plus your name, and you collect it with your ID. The data set at the retailer therefore looks just as harmless as with the pickup station, and you need no extra account. The price is opening hours and the occasional queue.

One variant is often overlooked in Germany: many carriers offer to redirect a shipment to a branch after the fact, as long as it is still in transit. That only helps so far, though, because by then the retailer has long since stored your home address. For the purpose at hand, all that counts is what goes into the address field during the order.

Buying in a shop creates no delivery data at all

The cleanest route is the oldest. Buy the device over the counter at an electronics market or a specialist crypto retailer and pay cash, and you leave no delivery address, because no delivery takes place. One second trade-off comes with it, however, and it has been argued over for years: the question of whether a device off the shelf could be tampered with.

The manufacturers answer that with procedures anyone can inspect. Reputable devices check at first start-up whether their software is genuine, and they require you to generate the recovery phrase yourself on the device. A device that presents you with a ready-made phrase on an enclosed slip of paper is to be disposed of in any case, whether it came from a shop or from a parcel. Take that check seriously at first set-up and buying in a shop brings no additional danger into the house, while saving you one data set.

A software wallet needs no shipping but costs security

The fourth route bypasses the parcel altogether. A software wallet on the phone or the computer is downloaded, not delivered, and therefore produces no shipping data. For smaller amounts, for first steps and for balances used to pay regularly, that is a defensible solution.

The comparison stays honest only with the other side of it. The key there sits on a device that is online, opens files and runs extensions. Balances have disappeared through exactly such extensions over the past months. The choice between the two forms is therefore not a choice between safe and unsafe, but between two different risks: with a delivered device you risk your address becoming known. With the downloaded application you risk your computer being compromised. Anyone holding larger amounts usually accepts the first risk and shrinks it through the ordering route.

What the ordering route does not solve: payment data, invoice and warranty

A pickup station does not make you invisible. Three things stay with the retailer even then. The billing address, which with most payment methods is checked against the payment provider's data and therefore has to be correct. The email address, through which order confirmation and delivery status run. And the payment trail itself, from which name and bank can be derived depending on the method.

In practice that means this: if you do not want to hand over the billing address either, buying in a shop is the only way. If you can live with your name and billing address sitting at the retailer while the delivery address does not turn up in every logistics database, route one or two serves you well. Warranty and the right of withdrawal, incidentally, hang on none of these decisions; they apply regardless of where the goods went.

Your name is already out there: access requests under Article 15 GDPR

For everyone who ordered before October, the ordering route comes too late. Data protection law applies instead, and it applies to manufacturers outside Europe too, as long as they deliver to customers in the European Union. Two duties fall on the company of its own accord: under Article 33 of the General Data Protection Regulation it must report a data breach to the competent supervisory authority within 72 hours of becoming aware of it. Under Article 34 it must also notify those affected without delay once there is a high risk to their rights, in clear and plain language, with details of the nature of the breach, the likely consequences and the measures taken.

Added to that is a right you can exercise yourself. The right of access under Article 15 obliges the retailer to tell you which data it processes about you, where it got them, who it passes them to and how long it keeps them. That third question is the most interesting one after the recent cases, because it brings to light the chain of service providers never mentioned during the order. An informal email to the address in the legal notice is enough, and a one-month deadline is the rule.

Deletion and damages: what Article 17 and Article 82 deliver

The next step follows from the access request. Anyone who is no longer a customer can demand deletion of their data under Article 17; retention duties under commercial and tax law take precedence and concern invoices above all, not every marketing list. The cases of recent months show why that is more than paperwork. At Keystone, buyers came forward whose order dated back more than two years, although the company had held out the prospect of deletion after 180 days in its privacy policy. At Trezor, old orders from the years 2019 to 2021 sat in the stolen files, which according to the manufacturer should contractually have vanished long before.

That leaves the question of money. Article 82 gives those affected a claim to compensation for non-material damage as well. On November 18, 2024 the German Federal Court of Justice ruled, on a data scrape at Facebook, that even the mere and brief loss of control over one's own personal data can constitute such damage; the court does not require proven misuse or further tangible consequences for that. It did not name a fixed sum, however, and the loss of control has to be set out in the individual case. If you are considering it, keep the retailer's notification email, because it is the simplest evidence that your data sat in the affected set.

52 recorded attacks in the first half-year: the physical risk in numbers

Why a delivery address weighs more heavily than an email address can be quantified. The analytics firm CertiK counted 52 confirmed physical attacks on holders of crypto assets in the first half of 2026, after 39 in the same period a year earlier. Recorded financial exposure rose, by the same analysis, from $10.5 million to around $124.1 million. The distribution stands out: 39 of the 52 cases fell on Europe, 33 on France alone. The method has shifted too, with attacks inside the home rising from one recorded case to 20 and abductions from 12 to 16.

These figures need context, and CertiK supplies it itself. The total includes ransom demands and forced transfers as well as frozen or recovered assets, so it is not a statement about actual loot. At the same time the firm points out that such crimes are heavily under-recorded. For judging a leaked address list the core holds: whoever owns a list of households that with some probability hold crypto assets owns a shortlist. As a countermeasure CertiK recommends, among other things, splits in which a single person cannot dispose of the balance alone, such as a multisig wallet with several keys, plus delays on withdrawals and clear arrangements within the family.

The trick after every leak: fake recall and replacement-device emails

The most common consequence of an address leak is less spectacular and hits more people. Keystone itself warned that fraudsters pose as employees of the company and offer refunds or replacement devices. The pattern has been the same in this industry for years: a message arrives that describes the real incident correctly, names you properly, often names the model ordered too, and derives an urgent offer from it. A free replacement device, a supposed security update, a compensation form.

The counter-rule is short and admits no exception. No manufacturer and no retailer ever asks for your recovery phrase, not by email, not on the phone, not through a form, not for verification and not for a refund. Enter the words and you lose your balance the moment you hit send, finally, because a blockchain transfer cannot be recalled. A device sent to you unasked should be set aside unused and reported to the manufacturer. How such messages can be spotted in detail is in our guide to phishing after a data breach.

Ordering a wallet: the parcel locker costs no extra

Protecting your keys and protecting your address are two different jobs, and no device does the second one for you. Three steps are enough.

1. Change the delivery destination before you order the next device. A pickup station or a branch as the delivery address costs nothing and takes its most valuable data set away from the longest chain. Which device suits your holding period and your amount at all is in the hardware wallet comparison.
2. Skip shipping entirely for small amounts. What you move day to day needs no parcel. Which applications are up to it and where their limits lie is shown by the software wallet comparison.
3. Do not tie everything to one known address. Part of the holding at a supervised provider reduces what sits at home and brings a different risk you should know about; which houses in Europe hold a licence is in the overview of regulated crypto exchanges.

(As of October 26, 2026. This article is not investment advice. Prices and fee structures change; check the terms with the provider before you buy.)

The facts: Polkadot trades at $1.006, or 0.899 euros, on Monday evening, October 26, down 3.3 percent in 24 hours (CoinGecko data, as of 18:52 UTC). Our assessment follows further down: the supply cap agreed in March is too small to carry the day-to-day price.

That leaves the price at a round number which last mattered in September. If you hold DOT, the long-term picture is in our Polkadot overview; this piece takes apart today's state of play and names the levels that count now.

Polkadot at the $1 mark: Monday evening's numbers

The day's range ran from $1.003 to $1.039. Tokens worth around $126.7 million changed hands over the past 24 hours. Market capitalization stands at about $1.72 billion, rank 57 among all crypto assets.

Over seven days DOT is down 11.9 percent, over 30 days 20.3 percent, over twelve months 68.2 percent. For comparison, bitcoin gave up 0.7 percent over the same 24 hours and stood at $80,024, while ethereum lost 1.6 percent to $2,315. DOT is falling roughly four times as hard as bitcoin this Monday.

The intraday low of $1.003 is the lowest level since September 16. DOT closed at $0.944 back then, and that was the last daily close beneath the round number. There have been 40 trading days above a dollar since, the most recent of them today, though only just.

Five straight losing sessions: what has changed since October 23

On October 23 we wrote here about the rebuild of the relay chain towards JAM. DOT closed at $1.1113 that day. The price has shed 9.5 percent since then, with nothing changed in the technical picture.

Last week's daily closes form an unbroken run lower: October 21 at $1.1505, October 22 at $1.1442, October 23 at $1.1113, October 24 at $1.0847, October 25 at $1.048, today at $1.006. Five trading days, each below the one before, a decline of 12.6 percent in total.

Part of that has a nameable cause. On October 21 the Polkadot Community Foundation warned about four tampered browser versions of Fearless Wallet, from which 17,286 DOT drained according to the foundation. We described the key leak at Fearless Wallet in detail at the time. The sum itself is too small to explain a slump in the price: at today's level, 17,286 DOT comes to around $17,400, or 0.014 percent of daily turnover. Trust is what moves the price here, not the quantity.

The 2.1 billion DOT supply cap and what it really slows

Polkadot changed its issuance model in March this year. Referendum 1710 replaced the fixed annual issuance with a stepped plan: every two years, 13.14 percent of the remaining gap to the ceiling of 2.1 billion DOT is issued anew, and the first step fell on March 14, 2026. That is how the official Polkadot wiki describes it. We reported on the hard ceiling at length on October 17, when only 392 million DOT could still come into existence.

Issuance schedule means the timetable on which new tokens are created. With bitcoin that is the halving, which halves issuance every four years. Polkadot pays out a fixed percentage of the remaining gap instead, so the annual quantity shrinks of its own accord.

How a cut in issuance reaches the price: the mechanism in three sentences

A cut in supply works through the ratio of new supply to current demand. If fewer new tokens reach the market each day, less buying power is needed for the price to hold. What decides that is not the ceiling as a number but the quantity actually created per day.

The calculation can be laid out. 1.709 billion DOT are in circulation, so the gap to the ceiling runs to around 391 million. 13.14 percent of that works out at about 51.4 million DOT for the two-year step, or roughly 25.7 million DOT a year. At $1.006 that is just under $25.9 million of new supply a year.

Set against it is daily turnover of $126.7 million. A whole year of new issuance therefore amounts to about a fifth of what changes hands on a single Monday. Cutting that quantity cannot hold the price against a wave of selling; it works over years, not weeks. That is precisely what the chart has shown since March.

The vote on Kusama on JAM Prime and the November 3 deadline

One date sits in the calendar, and it concerns DOT directly. On October 19 the group of authors around forum member Emielsebastiaan put four wish-for-change proposals for Kusama, the testnet with a real market value, to the Polkadot forum. Each proposal offers 40 coretime years but differs in the number of guaranteed cores: eight cores for five years, six cores for six years and eight months, four cores for ten years, two cores for twenty years. The paper names November 3, 2026 as the closing date.

One number is on the table for DOT holders. The funding is to be settled in DOT, with a negotiating basis of $4,320,000 and a ceiling of 5,000,000 DOT from the Kusama Vision Fund. Should one of the four variants fail, the result passes on to the DOT DAO through a proposal, according to the paper.

The proposals are forum papers so far. A query of Kusama referenda 671 to 674 through SubSquare returned no entry on Monday evening, so the votes had not been submitted at that point. To follow the deadline, watch the referenda list from this week on, not the forum.

Our assessment: the cap only counts once demand picks up

In the editorial team's view the supply cap is a sound decision carrying the wrong weight of expectation. The calculation above argues for that reading: $25.9 million of new supply a year against $126.7 million of turnover in a single day. So does the price since the first step on March 14, over which DOT has lost 68.2 percent on a twelve-month view. A rule about quantity is no substitute for demand.

The time axis argues the other way. The effect of such a plan adds up over several two-year steps, and the second stage is not due until March 2028. Hold DOT with a horizon of years and you are buying a scarcity that does not bite yet. Count in weeks and it will not be delivered. This is an assessment of the situation and not a recommendation to buy or sell; with crypto assets a total loss is possible.

What 28 days of lock-up in DOT staking mean now

Stake DOT and the tokens are not within reach straight away. The unbonding period on the Polkadot network runs to 28 days; only after that are the tokens transferable again. We worked through the yield and the lock-up on October 11, at 2.77 percent a year at the time.

On a day like today that period is the real decision. Unbond now and the earliest sale is November 23, and nobody knows where the price will stand then. Stay staked and the price risk runs on, but the yield sits above what a savings account leaves once inflation is deducted. Our comparison of staking platforms gives an overview of providers and terms.

On Wednesday, October 28, the US Federal Reserve decides on its key rate at 19:00 German time. That date falls in the middle of the 28 days if the unbonding starts today. It is not an argument against unbonding, but it is one for choosing the order deliberately.

Holding period and loss offsetting with DOT in the red

For investors in Germany, crypto assets held privately fall under the one-year holding period in Section 23 of the Income Tax Act. Hold DOT for longer than a year and the sale is tax-free, whatever the size of the gain. Sell within the year and the gain is taxed at your personal rate once the total of all private disposals in a year exceeds the threshold of 1,000 euros.

Loss offsetting means that losses from private disposals can be set against gains of the same kind, in the same year or carried forward into later years. With DOT down 68.2 percent over twelve months, that is the part which matters in practice for many portfolios.

The catch lies in the sequence. A loss from a sale after more than a year cannot be used for tax, because the transaction falls outside the period. To realize losses, then, check the acquisition date of each individual position first. The FIFO rule requires that the tokens bought first count as the ones sold first. Which tools read that out automatically is in our comparison of tax tools.

Levels above and below: $0.94 and the all-time low

Below, the next orientation is the daily close of September 16 at $0.944, around 6.2 percent under today's level. Beneath that follows the all-time low of August 18 at $0.727, a distance of 27.8 percent. No other notable close of the past 90 days sits between the two.

Above, yesterday's close at $1.048 counts first, then that of October 23 at $1.1113. The high of the past 90 days dates from September 28 and sits at $1.2677; DOT is 20.6 percent away from it today. These levels are reference points taken from the price history, not a forecast.

To buy DOT or shift holdings, the terms of the large venues are in our exchange comparison. With daily turnover of $126.7 million across all venues, the spread plays a measurable part in larger orders.

DOT at $1.00: without fresh demand the cap stays theory

Three steps follow from the situation:

1. Look up the acquisition date of every DOT position. If the purchase is less than a year old, a loss can be used for tax; after that it cannot. A tax tool reads this out of the transaction data.
2. Allow for the 28 days in staking. An unbonding started today ends on November 23, with the Fed date on October 28 in between. Current terms are in the staking comparison.
3. Mark November 3 in the calendar. The deadline on the four Kusama proposals runs out that day, with up to 5 million DOT of funding on the table.

(As of October 26, 2026. This article is not investment advice. Prices and fee structures change; check the terms with the provider before you buy.)

Whether a crypto gain triggers contributions to German statutory health insurance depends not on the size of the gain but on your insurance status. As a compulsorily insured employee you pay not a single cent to the fund on a gain from selling Bitcoin. As a voluntary member you do pay a contribution on it. And if you are covered by family insurance, a single sale can cost you free co-insurance for an entire year. These three routes sit in three different sections of the law, and reading the wrong one means calculating with the wrong number.

Right now the question is becoming concrete: on October 7, 2026 the German federal cabinet adopted the ordinance on social insurance parameters for 2027. The reference value rises to 4,130 euros a month, and because the income limit for family insurance is exactly one seventh of that, it will stand at 590 euros a month in 2027 instead of 565 euros. Approval by the Bundesrat is still outstanding, so the figures are not final. That changes nothing about the mechanics, and the mechanics are precisely why a crypto gain appears here at all.

Three groups of insured members, three completely different outcomes

Statutory health insurance has no single definition of income. Which receipts cost contributions is set out separately for each group in the Fifth Book of the German Social Code. For compulsorily insured employees, section 226 SGB V applies; for compulsorily insured pensioners, section 237 SGB V; for voluntary members, section 240 SGB V. And whether family insurance can exist at all is decided by section 10 SGB V on the basis of so-called total income.

These four provisions lead to three different outcomes for an identical crypto gain. It is therefore worth looking at each one before you write a number into a form.

Section 226 SGB V: for compulsorily insured employees the gain stays contribution-free

Section 226 paragraph 1 SGB V lists four items that form the basis of contribution assessment for people in compulsorily insurable employment: pay from the compulsorily insurable employment, the payment amount of the statutory pension, the payment amount of pension-equivalent benefits, and income from self-employment to the extent that it is earned alongside a pension or pension-equivalent benefits.

This list is exhaustive. A gain from a private disposal transaction does not appear in it. Income from self-employment, under section 15 SGB IV, means profit from self-employed activity, and a private sale of Ethereum from your own holdings is not self-employed activity. Employees, and with them the compulsorily insured, pay income tax on crypto gains but no health insurance contribution. The same applies to rental income, dividends and interest. That is the part that often comes as a surprise: the largest group of insured members is not affected by the contribution question at all.

Section 237 SGB V: what counts for contribution assessment in pensioners' health insurance

For compulsorily insured pensioners, that is health insurance for pensioners, section 237 SGB V names three items: the payment amount of the statutory pension, the payment amount of pension-equivalent receipts, and income from self-employment. Here, too, private disposal gains are missing.

A pensioner who sells after fourteen months of holding and makes a gain of 20,000 euros pays no contribution on it under this provision. What matters, though, is the type of insurance: a pensioner who is a voluntary member, and therefore not compulsorily insured, falls under section 240 SGB V and thus under an entirely different rule. It pays to know that status before you start calculating, because it appears in the notice from your fund and not in your pension notice.

Section 240 SGB V: the full economic capacity of voluntary members

For voluntary members the logic reverses. Section 240 paragraph 1 SGB V provides that contribution assessment is regulated uniformly by the national association of health insurance funds, and that this must ensure that the contribution burden takes account of the member's full economic capacity. Under section 240 paragraph 2 sentence 1, the receipts to be applied are at least those that would count for a comparable compulsorily insured person. At least, not at most.

This broad definition is the reason why voluntarily insured members, among them the self-employed, freelancers and voluntarily insured early retirees, have to declare investment income and disposal gains. The assessment basis is the gain as determined for tax purposes and not the sale proceeds: selling coins for 30,000 euros that cost 24,000 euros means a gain of 6,000 euros, not 30,000. At the upper end, the contribution assessment ceiling for health insurance applies, which is set to stand at 6,375 euros a month in 2027. Members who already reach that ceiling with their current income effectively pay nothing more on the crypto gain, because the maximum contribution is already in place.

Maximum contribution without proof: the legal consequence under section 240 paragraph 1 sentence 2

The harshest rule in contribution law sits in the second sentence of the same provision. If a voluntary member fails to submit the proof of contributory receipts requested by the fund, the contributory receipt deemed to apply for each calendar day is one thirtieth of the monthly contribution assessment ceiling. In plain terms: the fund sets the maximum contribution for as long as the proof is missing. That is not a penalty but a statutory presumption, and it takes effect without further examination.

There is a way back. Section 240 paragraph 1 sentence 3 SGB V gives you twelve months from notification of the assessment to apply for a new assessment. Members who then document their actual receipts have the contributions for the affected periods recalculated. Let the twelve months pass, and the maximum contribution stands. For that proof you need clean documentation of your purchases and sales across every platform, and that is exactly what the crypto tax software and portfolio trackers in comparison are for.

The one-seventh limit for family insurance rises to 590 euros in 2027

For spouses, registered partners and children without income of their own there is free family insurance under section 10 SGB V. One of its conditions is an income limit: the dependant must not have total income that regularly exceeds one seventh of the monthly reference value. Under section 18 SGB IV, the reference value is the average pay in the statutory pension insurance scheme in the calendar year before last, rounded up to the next higher amount divisible by 420.

For 2027 it is set to stand at 4,130 euros a month, after wage growth of 4.38 percent in 2025. One seventh of that comes to exactly 590 euros. For anyone in marginal employment, the higher marginal earnings threshold applies instead under section 10 paragraph 1 sentence 1 number 5 SGB V, which tracks the minimum wage. The exact figures will only be fixed once the ordinance is promulgated, because the Bundesrat still has to approve it.

What matters is the legal consequence when the limit is breached: family insurance does not exist for that period. The dependant has to insure themselves, usually as a voluntary member, and then section 240 SGB V with its full economic capacity applies once again. A contribution of zero can quickly turn into monthly contributions in the hundreds, and retroactively for the months in which the condition was not met.

Section 16 SGB IV: private disposal transactions form part of total income

That raises the decisive question: does a crypto gain count towards this total income at all? The answer follows from section 16 SGB IV. Total income is defined there as the sum of income within the meaning of income tax law. The national association of statutory health insurance funds has published basic guidance on this that the funds apply, and its point 2.3.4 explicitly names what belongs to other income under section 22 of the Income Tax Act: among other things income from annuities, from maintenance payments and income from private disposal transactions.

That is exactly where the crypto gain lands. Section 22 number 2 of the Income Tax Act lists income from private disposal transactions within the meaning of section 23, and section 23 paragraph 1 sentence 1 number 2 covers disposal transactions in other assets where not more than one year lies between acquisition and disposal. A taxable crypto gain is therefore other income and flows into the sum of income that forms total income.

The guidance is particularly strict in one place: for establishing total income, no reference may be made to the total amount of income, to income as defined after deductions, or to taxable income. Special expenses such as pension contributions and pure tax allowances are not deducted. The Federal Social Court confirmed this in 2004. The number that counts is therefore higher than what you read at the end of your tax assessment.

Regular or one-off: the sentence everything turns on here

The income limit applies to total income that is earned regularly on a monthly basis. Point 2.8 of the basic guidance explains how that is measured: a forward-looking approach is generally called for, and that requires a forecast taking into account the income circumstances that can be expected with sufficient certainty. Income received monthly is to be taken into account first, along with further regular income that is not received monthly but relates to the month.

And then comes the sentence that decides many cases: one-off receipts whose payment can be expected with sufficient certainty at least once a year are to be taken into account proportionately with the amount attributable to the month. The guidance names Christmas bonuses, holiday pay and additional monthly salaries as examples. The decisive feature is the characteristic of sufficient certainty.

For crypto investors that means: a single, unforeseeable sale is something other than a gain that reliably accrues every year. Someone who trades continuously and realises gains in every calendar year produces a forecast that carries those gains with it. A seller acting once, for a particular reason, argues differently. That classification is ultimately made by your health insurance fund, not by you, and in practice it is the point that gets disputed. So get the fund's assessment in writing before you trigger a large sale.

Worked example: a 4,200-euro crypto gain alongside a mini-job

The basic guidance annualises one-off receipts and divides the sum by twelve. That approach can be applied to a crypto gain. Assume a spouse is covered by family insurance, earns 470 euros a month in a mini-job and realises a crypto gain of 4,200 euros in June after a holding period of seven months.

The calculation runs like this: 470 euros times twelve comes to 5,640 euros of pay a year. Add the 4,200 euros of gain and the total is 9,840 euros. One twelfth of that is 820 euros a month. That lies above the limit of 590 euros, and above the higher mini-job threshold as well. Were the fund to include the gain in its forecast, family insurance would not be available for the year concerned.

The same calculation with a gain of 1,200 euros leads to 6,840 euros a year and 570 euros a month. That would keep the limit intact. Between the two cases lie 3,000 euros of gain and a contribution liability of several hundred euros a month. The tax return knows nothing of this threshold, because there it is only the tax rate that matters. How and where to declare the gains in the first place is set out in our guide to where gains, losses and staking income are entered.

Offsetting losses: what may enter the sum of income and what may not

After the price slide of the past few days this is the most practically important section for many readers. Bitcoin hovered around the 80,000-dollar mark on Monday midday, after the price had fallen in the morning to its lowest level since September. Selling at a loss in this environment also changes total income.

For the sum of income, the basic guidance permits the offsetting of positive and negative income within the same category of income, the horizontal loss offset. If losses remain after that, offsetting against gains from other categories of income is also possible, the vertical loss offset. An exception applies to investment income, where section 20 paragraph 6 of the Income Tax Act blocks the vertical offset.

The hard limit sits elsewhere: losses under section 10d of the Income Tax Act, that is the loss carryforward and carryback across calendar years, may not be deducted from the sum of income and are disregarded for total income. A pot of losses from the previous year helps you for tax purposes but not with your health insurance fund. For the tax side of those pots it is worth looking at what happens to older losses.

The 1,000-euro exemption limit and the one-year holding period

Two tax rules carry through into social insurance, because total income is tied to income as determined for tax. The first is the exemption limit: under section 23 paragraph 3 of the Income Tax Act, gains remain tax free if the total gain from private disposal transactions in the calendar year came to less than 1,000 euros. That is an exemption limit and not an allowance, and the difference is harsh: at 999 euros everything stays tax free, at 1,000 euros the full amount becomes taxable.

The second is the holding period. If more than a year lies between acquisition and sale, the transaction does not fall under section 23 of the Income Tax Act, and no taxable other income arises. Coins held for longer than a year and then sold create no item that shows up in the sum of income.

That gives the most effective lever of all, and it is also the dullest: the holding period decides both sides, the tax and the social insurance contribution. Whether your fund judges it exactly that way in every individual case cannot be promised across the board here. Have the legal consequence for your specific case confirmed by the fund or by a tax adviser.

Staking and lending: other income under section 22 number 3 of the Income Tax Act

Income from staking and lending is not a disposal gain but income from services under section 22 number 3 of the Income Tax Act. A separate and much lower limit applies to it: under section 22 number 3 sentence 2, such income is not subject to income tax if it came to less than 256 euros in the calendar year. Staking income of 300 euros a year is therefore taxable other income, which likewise flows into the sum of income and thus into total income.

For those covered by family insurance this is the less conspicuous risk. Ongoing staking income is recurring by its nature, so it falls under regular income more readily than a one-off sale does. Arithmetically, 300 euros a year is only 25 euros a month, which breaches no limit on its own. Combined with a mini-job, though, it can tip the balance.

Income checks and the duty to report: the route via the tax assessment

The funds check the income of dependants covered by family insurance at regular intervals and request proof for it, usually the income tax assessment. Voluntary members submit it in any case, because contribution assessment under section 240 SGB V is built on it. A taxable crypto gain appears in that assessment under other income and is therefore visible to the fund.

Two practical duties follow from this. First: changes in income circumstances that affect family insurance are to be reported to the fund, and ideally before the next check. Second: a voluntary member who submits no proof risks the assessment of the maximum contribution, with only twelve months to correct it.

How crypto assets and income are treated in other social benefits follows separate rules of its own. For the field of basic income support we have written that up separately, namely how crypto assets are counted towards Bürgergeld. The limits there have nothing to do with the contribution rules of health insurance.

What this text cannot deliver

Two points remain open, and that is part of the picture. The parameters for 2027 have been adopted but not yet promulgated, because the Bundesrat has to approve them. Until then, a reference value of 4,130 euros and an income limit of 590 euros are expected values, not final ones.

And on the treatment of a one-off crypto gain in the income forecast for family insurance there is no published decision that settles the case expressly for crypto assets. What is documented is the legal framework: that private disposal transactions form part of total income and how one-off receipts are to be treated. How your fund classifies your individual case is something you have to ask it. A written answer before the sale is worth more than any interpretation afterwards.

Crypto and the health insurance fund: family insurance ends at 590 euros

The limit works quietly and hits hard, because it is tied to a number that appears in no price analysis. Three steps are worth taking before you trigger your next sale:

1. Establish your status. Check the notice from your fund to see whether you are compulsorily insured, a voluntary member or covered by family insurance. Only in the latter two cases is the gain relevant at all. For documentation across every platform, the crypto tax software and portfolio trackers in comparison will help.
2. Calculate the holding period and the annual total. Add up the realised gains of the calendar year, divide the sum including other income by twelve and compare it with your limit. Where you buy and sell, and which tax reports a platform provides, is shown in the overview of crypto exchanges in comparison.
3. Get an answer, in writing. Ask your fund before a large sale how it treats the gain in its income forecast. Falling back on verifiable records from licensed trading venues makes that easier, and which ones those are is set out under regulated crypto exchanges in comparison.

Sources: the press release of the German Federal Ministry of Labour and Social Affairs on the social insurance parameters for 2027 and the basic guidance of the National Association of Statutory Health Insurance Funds on total income.

(As of October 26, 2026. This article is not investment advice. Prices and fee structures change; check the terms with the provider before you buy.)

Here are the facts: the capital locked in DeFi applications on the Solana blockchain stands at $5.72 billion in DefiLlama's daily series, and at $5.75 billion when we pulled the figure on Monday afternoon. The last time the value was lower was September 16, 2026. Over the same period, $1.45 billion changed hands in 24 hours on the blockchain's decentralised exchanges, more than on Ethereum and Base combined. Our assessment follows further down: we treat this gap between shrinking capital and steady volume as the more telling signal than the price level, and we also name the figure that argues against it.

The price itself barely moved. CoinGecko put it at $100.89 on Monday afternoon, 0.45 percent below the level of 24 hours earlier, with a market capitalisation of $59.49 billion. Where the price can point over a horizon of months is set out in our Solana price prediction. For this Monday the price is the least telling number of the day, because it stayed quiet while something shifted underneath it.

$5.72 billion: Solana's DeFi capital is back at its mid-September level

Total value locked, or TVL, measures the dollar value of all deposits in a blockchain's applications: liquidity in trading pools, collateral in lending protocols, deposits in yield products. The figure says nothing about how much happens on a blockchain. It says how much money sits there and works.

On Solana that is currently $5.72 billion. On September 28, the high of the past 30 days, it was $6.61 billion. That is 13.4 percent less in four weeks. The decline did not come as a one-day drop but in steps: $5.98 billion on October 22, $5.86 billion on the 23rd, $5.77 billion on the 24th, $5.78 billion on the 25th and now $5.72 billion. Any single day looks like noise. The series does not.

Part of that is arithmetic rather than an outflow. TVL is measured in dollars, and when the SOL price falls, the same stock of tokens shrinks on paper with it. SOL currently trades 65.6 percent below its all-time high of $293.31. The other part is real money that investors and funds have pulled out of the applications. How large each share is cannot be separated from the dollar series alone, and anyone who claims otherwise has not measured it.

Base has been ahead since October 10

At the time of the Monday afternoon pull, Solana ranked third among blockchains by locked capital with $5.75 billion. Ethereum leads by a wide margin with $48.33 billion, Base follows with $6.03 billion, then BNB Chain with $5.52 billion and Tron with $5.44 billion.

The swap between Base and Solana is a good two weeks old. On October 10 both blockchains stood at around $6.16 billion in the daily series, and Base has been ahead since. At roughly $280 million the gap is small enough for one quiet week to erase it. It is still worth noting, because Base and Solana take entirely different routes: Base is a layer-2 blockchain on Ethereum with a strong tie to a single exchange, Solana a standalone blockchain with its own validator set.

$1.45 billion in daily volume: trading holds, capital does not

The second number tells a different story. Decentralised exchanges on Solana handled $1.45 billion over the past 24 hours. On Ethereum the figure was $726 million, on Base $716 million. Solana therefore turns over slightly more in a day than the two combined.

Over seven days the picture is the same: $13.44 billion on Solana, $7.56 billion on Ethereum, $6.03 billion on Base. Over 30 days Solana reaches $62.22 billion. Trading has not collapsed, then, while locked capital gave up 13.4 percent. That simultaneity is the finding of the day.

Capital turnover compared: Solana 25 percent, Base 12 percent, Ethereum 1.5 percent

Taken together, the two numbers produce a third that appears on no dashboard and that we therefore calculated ourselves: capital turnover. It divides daily volume by locked capital and says how often a deposited dollar is traded in a day.

• Solana: $1.45 billion in volume on $5.75 billion of capital, or 25.2 percent in a single day.
• Base: $716 million on $6.03 billion of capital, or 11.9 percent.
• Ethereum: $726 million on $48.33 billion of capital, or 1.5 percent.

A dollar of liquidity on Solana works about twice as often in a day as on Base and roughly seventeen times as often as on Ethereum. That is not a verdict on the blockchains but a description of how they are used: on Ethereum a lot of capital sits long term as collateral, on Solana comparatively little capital turns over very fast. For a forecast it means that Solana depends less than other blockchains on locked capital and more on live trading. Should trading fall away, the support goes faster than elsewhere.

How locked capital, trading volume and fees hang together

The mechanism behind it takes three sentences. Whoever puts liquidity into a trading pool earns a fee on every swap. The higher the volume for the same pool, the higher the return on the capital deposited, and the more likely that capital stays put. If volume falls while the capital remains, the return per dollar drops and the capital moves away.

On Solana the rarer direction is playing out right now: the capital leaves, the volume stays. Arithmetically that lifts the fee yield of the remaining liquidity providers, because the same volume is spread across fewer deposits. That is the stabilising part of the picture, and it is the reason a falling TVL alongside steady volume is not the same thing as a falling TVL alongside falling volume. Providers of liquidity, and anyone earning interest through lending protocols, will find the platforms and their terms in our crypto lending comparison.

The counter-test is still outstanding. Should volume also give way in the coming days, the reading flips: the TVL decline would then have been the precursor of a broader retreat and not the consequence of a price drop. That is visible in the seven-day volume series, not in a single day.

437.7 million SOL are staked, 152.1 million move freely

To understand how sensitive the price is to capital movements, it helps to know how much SOL is freely tradable at all. We measured this on Monday afternoon directly against a node of the Solana blockchain rather than taking it from a statistics page.

There are 589,819,536 SOL in circulation. Of those, 437,716,746 SOL are staked with validators, spread across 664 active and 26 delinquent validators. That is 74.21 percent of the circulating supply. Freely moving, therefore, are 152,102,790 SOL, worth roughly $15.3 billion at the current price.

This calculation has two sides. A high staking share takes supply out of the market and supports the price as long as holders stay the course. It also makes the freely traded remainder thinner, and a thin market moves further in both directions for the same amount of capital. Staked SOL is not immediately available either: undoing a delegation only takes effect at the end of the current epoch.

Inflation of 3.59 percent against a staking yield of 4.96 percent

New tokens keep being issued throughout. The inflation rate of the Solana blockchain stands at 3.5857 percent a year in the current epoch, measured on Monday afternoon through the network interface. The share going to the Solana Foundation is zero; the entire issuance flows to validators and to the accounts that have delegated SOL to them. How that path is laid out is described in the technical documentation on the inflation curve: the rate falls by a fixed percentage every year until it reaches a floor.

Against that stands the staking yield. Our article of October 25, 2026 put it at 4.96 percent a year; that was the return a delegated account could expect over a year. The difference to the inflation rate is around 1.4 percentage points. That is how much share of the total supply an account that stakes gains in a year over an account that merely holds the same amount of SOL. In the same calculation, the holder without staking loses about 3.5 percent of their share of the total supply in a year, even though their token count stays the same.

Both figures move. The yield depends on how much SOL is staked in total and how high fee income turns out; the inflation rate follows the fixed path and falls with every epoch-year boundary. Calculations based on expected values are therefore better built on the difference than on the yield alone.

Epoch 1069 starts in the night to Tuesday

Solana counts in epochs of 432,000 slots. The current epoch carries the number 1068 and stood at slot 360,434 on Monday afternoon, or around 83 percent. At a target time of about 0.4 seconds per slot, just under eight hours remain until the switch to epoch 1069, which therefore falls in the night to Tuesday.

The epoch change is not a price event, but it is a date with practical consequences. At that moment staking rewards are distributed, newly delegated SOL becomes active, and undone delegations become freely available. Anyone who cancelled a delegation in recent days has their SOL back to hand from the switch onwards. In total this regularly involves sums in the hundreds of millions that can reach the market, though they need not.

Day low $99.82, day high $102.48

The daily range was narrow. Between the low at $99.82 and the high at $102.48 lie $2.66, or 2.64 percent of the day's low. The 24-hour volume across all recorded venues stood at $2.25 billion according to CoinGecko.

Three values can be justified from that range. The round $100 level sits in the middle of the daily range and was crossed several times in both directions today; it therefore serves as orientation, not as a signal. The day low at $99.82 is the lowest point at which buyers stood today. The day high at $102.48 marks the point where they ran out of air. Traders working with leverage calculate their liquidation distance against those two values and not against a round number.

Compared with our last piece on the Solana price prediction, the picture has shifted slightly. On October 25, when we looked at the USDC rollout to 82 million Samsung devices, SOL stood a good two dollars higher, and the level discussed then has held since without turning into upward movement. What is new is the TVL decline, which at that point was not yet a six-week low.

Holding period and staking: the rule for investors in Germany

This is where the German situation parts from international coverage. Gains from selling SOL count as a private disposal transaction under section 23 of the German Income Tax Act. If more than twelve months lie between purchase and sale, the gain remains tax free. Below that it counts towards taxable income as soon as the total of all private disposal transactions in a year reaches the exemption limit of 1,000 euros. An exemption limit is not an allowance: one euro above it makes the entire amount taxable.

For staking, a clarification has applied since the German Federal Ministry of Finance's letter of May 10, 2022 that many investors have missed to this day: staking does not extend the holding period to ten years. It stays at twelve months. The ongoing staking rewards themselves are untouched by that; they count as other income under section 22 number 3 and are to be recognised at their value at the time of receipt, with an exemption limit of their own of 256 euros a year. For rewards that are sold, a separate twelve-month period then runs from receipt.

In practice that means: holders who have held and staked SOL since autumn 2025 have already completed the period for the original position, while every single reward has its own clock running. Tracking that by hand quickly becomes unmanageable with weekly inflows. Tools that record the time of receipt automatically take that work off your hands; the providers are linked further down. This section does not replace tax advice; for larger amounts the case belongs in expert hands.

Our assessment of the situation at Solana

In the view of the newsroom, the TVL decline on its own is the weaker signal and its combination with steady volume the stronger one. Three documented figures support that: volume of $13.44 billion over seven days, which exceeds that of Ethereum and Base; capital turnover of 25.2 percent a day, which is more than double the Base figure; and a staking share of 74.21 percent, which keeps the freely traded supply at around $15.3 billion. A blockchain whose usage stays stable while the money parked there shrinks loses buffer, not demand.

Against that stands a figure we are not going to explain away: a 13.4 percent decline in four weeks is not noise, and Base has been ahead for two weeks. Should the seven-day volume series also give way in the coming days, our reading collapses, and the TVL decline was the earlier indicator. The Fed rate decision on Wednesday is the nearest date at which direction for the broader market can be settled. This assessment evaluates the situation and is not a buy or sell recommendation.

Solana at $100.89: volume carries, capital does not

1. Start by comparing the seven-day series of trading volume with that of locked capital instead of looking only at the price level. If both fall, the situation changes fundamentally. Which platforms show these series side by side is set out in our comparison of analytics platforms.
2. If you hold SOL for longer than twelve months, note the purchase date and the receipt dates of staking rewards separately. That decides tax exemption, not the price. The tools for it are in the comparison of tax software.
3. Before buying, weigh the venue's fees against the expected holding period: for a position meant to sit for twelve months, a one-off trading fee weighs more than the daily price level. The terms offered by regulated providers are in the exchange comparison.

The figures on locked capital and trading volume come from DefiLlama, the price data from CoinGecko, and the values for supply, staking, epoch and inflation rate from a direct query of the Solana blockchain.

(As of October 26, 2026. This article is not investment advice. Prices and fee structures change; check the terms with the provider before you buy.)

Unitree's shares have jumped about 60% in under two weeks as investors bet OpenAI's GPT-6 Astra will give robots the brains they lack, turning the stock into an AI proxy.

The agreement, which follows similar deals with MLB and NHL, establishes information-sharing to guard against insider trading as sports betting migrates to CFTC-regulated venues.

Police in Seoul say a 32-year-old with no IT background used AI tools to break into about 90 hospitals, schools and companies, then pressed victims for roughly $270,000 combined.

Cathedral, a startup founded by ex-DOGE staffers, says its AI agents found paths into Anthropic's sales, HR, and IT systems through third-party software.

Experts say Brazil's prediction market ban left the odds in the hands of traders with little knowledge about the election.

XRP Ledger co-creator David Schwartz has shared an enthusiastic message from Ripple's Swell Demo Day in New York, fueling anticipation ahead of the company's flagship conference.

Billionaire venture capitalist Tim Draper has praised Bitcoin creator Satoshi Nakamoto for his bold vision and optimism, crediting the mysterious figure with challenging the foundations of the global financial system.

American Airlines, Subway and Christian Louboutin already use Ripple’s new 24/7 treasury platform, which puts idle cash to work via RLUSD.

Dogecoin faces a major ideological crisis after a new corporate merger transfers key developer accounts to a private firm.

BitMine is rapidly drawing closer to owning 5% of Ethereum's supply as its total Ethereum holdings are now worth over $14 billion.

TLDR

• Amkor Q3 revenue reaches $1.99 billion as gross margin expands to 18.3% in 2026.
• AMKR shares fall 6.34% in regular trading before gaining 3.23% after hours.
• Amkor’s Q3 net income rises to $191 million, with EPS increasing to $0.76 per share.
• Computing and automotive revenue set records despite softer communications demand.
• Amkor forecasts up to $2 billion in Q4 sales and $2.5 billion in annual capex.

Amkor Technology (AMKR) reported third-quarter revenue of $1.99 billion as stronger margins supported higher earnings across its semiconductor business. The stock fell 6.34% to $36.81 during regular trading before rebounding 3.23% to $38.00 in after-hours trading. Record computing and automotive revenue supported quarterly results despite weaker communications demand.

Amkor Technology, Inc., AMKR

Amkor Technology Reports Higher Revenue and Operating Margins

Amkor generated $1.991 billion in net sales, compared with $1.898 billion in the second quarter. Revenue also exceeded the $1.987 billion recorded during the corresponding quarter of 2025. The modest annual increase accompanied stronger gains in operating profitability.

Gross profit reached $364 million, while gross margin improved to 18.3% from 16.8% in the previous quarter. The company reported a 14.3% gross margin one year earlier, marking a four-percentage-point improvement. This expansion reflected stronger profitability despite relatively flat annual revenue.

Operating income climbed to $225 million from $200 million in the preceding quarter and $159 million last year. Meanwhile, operating margin increased to 11.3%, compared with 10.5% sequentially and 8.0% annually. These results reflected improved earnings across Amkor’s semiconductor packaging and testing operations.

Amkor Net Income Climbs as Computing Demand Strengthens

Net income attributable to Amkor reached $191 million, compared with $174 million during the previous quarter. The company earned $127 million during the third quarter of 2025. Diluted earnings increased to $0.76 per share from $0.70 sequentially and $0.51 annually.

EBITDA climbed to $435 million from $400 million in the second quarter. The figure also exceeded the $340 million reported during the corresponding period last year. Amkor attributed its performance to stronger customer programs, improved margins, and expanding business commitments.

Computing and Automotive & Industrial operations delivered record revenue during the reporting period. However, the Communications segment experienced expected weakness, limiting overall sales growth. Stronger customer commitments also improved demand visibility and supported the company’s investment planning.

Amkor Sets Fourth Quarter Outlook and Capital Spending Plans

Amkor expects fourth-quarter revenue between $1.9 billion and $2.0 billion. The company projects gross margins between 18.0% and 19.0% during the period. This outlook indicates continued margin strength heading into the final quarter of 2026.

Management forecasts fourth-quarter net income between $160 million and $185 million. The range translates into diluted earnings of $0.65 to $0.75 per share. Amkor also updated its full-year capital expenditure forecast to approximately $2.5 billion.

At September’s end, Amkor held $2.4 billion in cash and short-term investments against $2.5 billion in debt. The company repurchased approximately $32 million in shares during the third quarter. It also paid a quarterly dividend of $0.08352 per share on September 22, with future payments requiring board approval.

The post Amkor Technology (AMKR) Stock: Q3 Revenue Hits $1.99 Billion as Margins Expand appeared first on Blockonomi.

TLDR

• CLS stock rebounds 9.17% after hours, reversing a 5.14% regular-session drop.
• Celestica Q3 revenue surges 83% to $5.83 billion as cloud infrastructure expands.
• Adjusted earnings jump 109% to $3.30 per share, with margins reaching 8.5%.
• Celestica raises its 2026 revenue outlook to $21 billion and EPS to $11.55.
• Celestica targets $40 billion in 2027 revenue as AI infrastructure demand grows.

Celestica Inc. (CLS) shares surged after hours Monday after the company reported an 83% jump in third-quarter revenue. Sales reached $5.83 billion, while adjusted earnings per share more than doubled. CLS gained 9.17% to $268.99 after regular trading ended, reversing an earlier decline.

Celestica Inc., CLS

Celestica Q3 Earnings Drive After-Hours Rebound

During regular trading, Celestica stock fell 5.14% to $246.65, shedding $13.36 before its earnings announcement. The subsequent rally followed results for the quarter ending September 30, 2026. Revenue climbed from $3.19 billion in the comparable 2025 quarter.

Adjusted earnings reached $3.30 per diluted share, compared with $1.58 in the prior-year quarter. Meanwhile, GAAP earnings increased to $3.26 per diluted share from $2.31 a year earlier. Adjusted earnings per share grew 109% year over year.

Celestica reported an 8.5% adjusted operating margin, up from 7.6% in the comparable quarter. However, its GAAP operating margin declined to 7.9% from 10.2% last year. A completed public offering raised $3.45 billion in gross proceeds.

Celestica Raises 2026 Revenue and Earnings Forecasts

Following the quarterly results, Celestica increased its full-year revenue outlook to $21 billion from $20.5 billion. It also raised its adjusted earnings forecast to $11.55 per share from $11.30. Management linked the increases to recent results and stronger fourth-quarter demand.

For the fourth quarter, the company projected revenue between $6.275 billion and $6.575 billion. It expects adjusted earnings of $3.37 to $3.57 per diluted share during that period. Its guidance assumes an 8.7% adjusted operating margin at the forecast midpoint.

Celestica reduced its 2026 free cash flow forecast to $460 million from $600 million. The company attributed the revision to working capital investments supporting faster expansion next year. Its previous forecast for a full-year adjusted operating margin of 8.4% remains unchanged.

AI Infrastructure Demand Supports 2027 Expansion Plans

Celestica expects 2027 revenue to reach $40 billion, representing 90% growth over its updated 2026 target. The company also forecasts adjusted earnings of $21.50 per share for next year. It targets a 9.0% adjusted operating margin, which would set another company record.

Celestica plans capital spending between $1.5 billion and $2 billion during 2027. Its growth strategy centers on equipment and infrastructure for large cloud operators and advanced AI developers. Management also indicated that customer demand projections for 2028 exceed current expectations for 2027.

Celestica operates two segments: Connectivity and Cloud Solutions, and Advanced Technology Solutions. Its cloud business includes networking, servers, and storage products, while other operations serve aerospace, healthcare, and manufacturing. The updated forecasts reflect greater capacity needs as customers expand their orders.

The post Celestica Inc. (CLS) Stock: Jumps as Q3 Revenue Soars 83% to $5.83 Billion appeared first on Blockonomi.

TLDR

• SpaceX (SPCX) stock falls 0.40% to $146.90 before a 0.30% after-hours rebound.
• SpaceX secures $40 billion in GPU financing led by Apollo and Goldman Sachs.
• Nvidia Vera Rubin chips will expand Grok computing and data center capacity.
• SpaceX expects GPU leasing to begin in the first half of 2027 under its deal.
• Rising debt costs and GPU depreciation put pressure on specialized cloud firms.

Space Exploration Technologies Corp. (SPCX) stock declined 0.40% to $146.90, shedding $0.59 before recovering 0.30% to $147.34 after hours. The decline came as SpaceX announced a $40 billion financing agreement to acquire Nvidia GPUs for its computing operations. The agreement will support Grok development and expand the company’s data center infrastructure through its xAI business.

Space Exploration Technologies Corp., SPCX

SpaceX Secures $40 Billion Financing for Nvidia GPUs

SpaceX arranged the financing through a master finance lease involving Apollo Global Management and Goldman Sachs. Apollo and Goldman Sachs led the financing commitments, while Goldman Sachs served as the structuring agent. Affiliates of BlackRock, Goldman Sachs Asset Management, and KKR also participated in the transaction.

Under the agreement, an Apollo-managed entity will purchase Nvidia Vera Rubin GPUs and related equipment for SpaceX. SpaceX will lease the hardware through separate tranches, with each lease lasting approximately five to 5.25 years. The company will make quarterly payments covering principal and a fixed annual financing rate.

The financing structure also requires rent equal to 5% of each tranche’s financed cost. SpaceX expects to complete the agreement before year-end, with equipment leasing scheduled to begin during the first half of 2027. The company described the arrangement as the largest committed GPU financing transaction to date.

Grok Expansion Drives SpaceX Data Center Investment

SpaceX plans to deploy the Nvidia processors across infrastructure supporting Grok and other xAI computing requirements. The equipment will provide additional capacity for training models and processing requests as computing demand expands. Furthermore, the investment supports terrestrial data center development alongside the company’s longer-term orbital computing plans.

SpaceX absorbed xAI in February, combining its space operations with the artificial intelligence business. The integration brought Grok development and computing infrastructure under the company’s broader investment strategy. Meanwhile, Starlink subscriptions and rocket launches continue generating revenue outside the data center business.

Moody’s assigned SpaceX a Baa1 investment-grade rating in June, strengthening its position in credit markets. Elon Musk subsequently warned that lenders had restricted financing access for companies dependent on AI data center revenue. He also predicted consolidation among specialized cloud providers facing tighter borrowing conditions.

AI Financing Pressures Raise Questions About GPU Economics

The SpaceX agreement follows a $36 billion structured financing transaction supporting GPU purchases for Anthropic. Apollo and Blackstone led that earlier arrangement, which involved Broadcom and equipment leasing commitments. Consequently, large technology infrastructure projects continue attracting financing despite growing concerns about hardware values.

CoreWeave illustrates the financial pressure facing specialized GPU cloud providers as borrowing costs increase. Its shares declined approximately 30% between September 30 and October 16, while its debt approached $50 billion. Meanwhile, its unsecured bonds yielded approximately 15% in mid-October, reflecting expensive financing conditions.

GPU depreciation presents another challenge because computing hardware can lose value before financing agreements expire. CoreWeave recorded nearly $1.4 billion in depreciation and amortization against approximately $2.6 billion in second-quarter revenue. Venture capitalist Chamath Palihapitiya also expects unused computing capacity and industry consolidation during 2027.

The post SpaceX (SPCX) Stock: Drops as $40 Billion GPU Deal Boosts Grok and Data Center Plans appeared first on Blockonomi.

TLDR

• Vertex launches Sōl to connect tax reporting, audit records, and AI decisions.
• VERX stock slides 1.49% to $15.16 following the launch of its Vertex Sōl platform.
• Vertex Sōl gives tax teams greater control over AI recommendations and decisions.
• A Vertex survey finds 81% demand full visibility before adopting compliance AI.
• New tools link tax calculations, e-invoicing, reporting, and audit records.

Vertex, Inc. (VERX) has introduced Vertex Sōl, a new platform that connects tax compliance processes through artificial intelligence. The system combines tax decisions, regulatory intelligence, reporting, and audit records within a unified framework. Meanwhile, VERX stock fell 1.49% to $15.16, losing $0.23 after recovering from an intraday low near $14.87.

Vertex, Inc., VERX

The launch expands Vertex’s technology portfolio as businesses seek greater control over automated tax operations. Vertex Sōl connects compliance activities across individual transactions, from initial tax calculations through reporting and audits. Tax departments can identify potential errors earlier and manage supporting records through one system.

Vertex Sōl Brings AI Into Tax Compliance Operations

Vertex Sōl integrates tax determination, electronic invoicing, regulatory reporting, and audit preparation into a connected workflow. The platform maintains transaction records throughout each compliance stage, reducing the need for separate documentation processes. Furthermore, Vertex Intelligence supports the system with regulatory information, tax expertise, explanations, and recommendations.

The company designed the platform to give tax professionals direct control over automated decisions. Teams can review recommendations, approve actions, and reverse decisions when necessary. This structure allows businesses to adjust automation levels according to internal policies and compliance requirements.

Vertex serves a customer base that includes more than 60% of Fortune 500 companies. Its existing tax solutions support businesses that manage complex transactions across different regulatory environments. The new platform extends those services by connecting tax calculations with documentation and audit requirements.

Vertex Survey Highlights Demand for AI Oversight

Vertex introduced the platform as enterprises examine the risks and benefits of automated compliance systems. A company-commissioned survey questioned 402 tax and compliance software decision-makers across the United States and Germany. Emerald Research Group conducted the research to assess expectations surrounding artificial intelligence in compliance operations.

The survey found that 81% of respondents required full visibility into automated decisions before adopting AI compliance tools. Respondents also wanted the ability to override or reverse automated actions at any time. These findings highlight the importance of transparency and human control in regulated financial operations.

Vertex Sōl addresses these requirements through adjustable automation settings and decision records. Tax departments can choose whether the system recommends actions, assists employees, or operates within established controls. Meanwhile, its recordkeeping functions allow teams to trace decisions back to their original transactions.

Vertex Expands Compliance Capabilities Across Tax Workflows

Vertex Sōl supports several activities that tax departments handle throughout the reporting cycle. When regulations change, the platform connects updated requirements with relevant transactions and identifies potential compliance impacts. Tax professionals can then review the information before making operational changes.

Before filing, the system identifies relevant issues and provides recommendations within company-defined controls. During audits, it helps teams organize notices, gather supporting documents, and identify missing information. These functions connect routine compliance work with the evidence businesses need during regulatory reviews.

Industry research firm IDC identifies governance and traceability as important requirements for enterprise compliance technology. Its assessment emphasizes the need to explain automated actions and connect them with original tax decisions. Vertex’s latest launch reflects that direction while extending its existing compliance software capabilities.

The post Vertex, Inc. (VERX) Stock: Unveils Vertex Sōl to Transform Tax Compliance With AI appeared first on Blockonomi.

TLDR:

• Santiment counted 59.95 million non-empty Bitcoin addresses on October 26, down 30,000 since October 16.
• BTC touched $79,332 on Monday before trading near $80,000, with resistance marked at $82,000–$82,500.
• Ali Martinez flagged a four-hour channel breakdown, identifying $72,500 as a potential buying area for BTC.
• U.S. spot Bitcoin ETFs lost $635.6 million on October 22 and $137.9 million on October 23, Farside said.

Bitcoin’s loss of 30,000 funded wallets has created a market paradox, as declining retail participation coincides with mounting warnings of another selloff. Consequently, the contrasting signals raise questions about whether investors are witnessing seller exhaustion that could establish a market bottom or continued weakness pointing to further losses.

Santiment reported October 26 that Bitcoin had approximately 59.95 million non-empty addresses, down 30,000 since October 16. Its findings suggest discouraged smaller holders may be withdrawing after prolonged stagnation. The count measures funded blockchain addresses, rather than individuals, making it an indirect indicator of shifts in retail behavior.

Yet BTC traded near $80,000 on Monday after touching roughly $79,332. That decline leaves price analysis confronting two competing readings: seller exhaustion and a weakening technical structure. BTC must also contend with resistance around $82,000, while another chart assessment identifies a buying area substantially lower.

30,000 Bitcoin Wallets Lost as $80,000 Bottom Faces Test

Santiment describes shrinking retail participation as a potential contrarian signal. Historically, recoveries can follow periods when impatient holders abandon positions, reducing speculative enthusiasm. However, a wallet is not an investor.

One person may control several addresses, meaning Bitcoin address losses cannot establish precisely how many people sold. Santiment’s October 9 price analysis showed the opposite trend.

At that point, addresses holding less than 0.01 BTC were increasing despite weaker markets. The firm then viewed continued retail buying as an obstacle to recovery. By October 26, its price analysis highlighted falling wallet numbers instead. For BTC, the immediate reference is $80,000.

Stabilization there, followed by a move beyond $82,500, would strengthen the firm’s proposed bottoming interpretation. Conversely, a sustained break below $79,000 would leave that interpretation unsupported. Bitcoin still needs observable buying demand rather than address counts alone.

BTC Channel Breakdown Puts $72,500 in Focus Amid ETF Outflows

Beyond the on-chain signals, technical indicators reinforce concerns about further downside. Ali Martinez’s price analysis shows BTC breaking below a rising channel on the four-hour chart, disrupting its previous pattern of higher lows.

Consequently, Martinez identifies $82,000 as resistance and $72,500 as a potential buying area, approximately 9% below Monday’s Bitcoin price. Meanwhile, institutional positioning adds another dimension to the market’s uncertainty.

Glassnode reported on October 21 that U.S. spot Bitcoin ETF holders had an average entry price near $81,500. Corporate treasuries, by comparison, held an average purchase price around $80,200.

Furthermore, Glassnode’s price analysis identified the True Market Mean near $77,600, providing another reference below $80,000 without confirming further losses. Recent institutional flows also highlight the challenges facing a sustained recovery.

According to Farside Investors, spot Bitcoin ETFs recorded $635.6 million in outflows on October 22, followed by another $137.9 million on October 23. Against this backdrop, BTC price analysis centers on reclaiming $82,000–$82,500 or extending losses below $79,000.

Adding to the uncertainty, the Federal Reserve’s October 28 policy decision presents another near-term market catalyst. Until resistance is reclaimed, this price analysis leaves Bitcoin without a confirmed bottom, while BTC remains exposed to the lower support levels identified by analysts.

The post Bitcoin Bottom or Bull Trap? 30,000 Wallets Exit as $72,500 Risk Grows appeared first on Blockonomi.

Ethereum co-founder Vitalik Buterin said in a post on X on Monday that he is comfortable with the AI model Kimi 3.1 deliberately leaving top-end offensive cyber capabilities out of its training data.

In his view, the trade-off is worth making because it gives defenders more time to prepare.

Restricting Use Cases, Not Users

“Discriminating against use cases is MUCH less bad than discriminating against users,” Buterin wrote.

The world, he added, needs time to harden systems and check old code before “the full weight of very powerful cyber attacks is upon us.” In plain terms, he prefers a model that declines certain tasks for everyone over one that turns certain people away.

He also sketched what could go wrong. If “this first big wave of AI risk is not handled well,” he cautioned, the result might be a “dystopia of highly locked down models,” with KYC checks, which tie an account to a verified identity, required for everything and intrusive surveillance to match. Depending on credentials and nationality, a person could end up a “second or third or fourth-class citizen” in terms of model access.

Asked by X user LOID whether other labs would build those capabilities anyway, Buterin replied: “Yes of course they will.” Read alongside that argument about time, the filtering is meant to delay such capabilities rather than prevent them.

Not everyone agreed. “Hard disagree,” wrote developer Youssof Al Toukhi, who brought up the recent attack on Hugging Face by an OpenAI model, and when Hugging Face tried to analyze the attackers’ actions using frontier models, the providers’ safety guardrails blocked the requests, forcing a switch to an open model, GLM 5.2. Calling the arrangement an “elitism system” that “will only get worse overtime,” the critic added: “Intelligence needs to be democratized.”

Earlier Remarks on AI and Code Security

As CryptoPotato reported previously, Buterin countered fears in May that AI-powered bug hunting would overwhelm developers and create non-stop openings for exploits, arguing that AI-assisted formal verification could become one of the strongest defenses in crypto.

In his telling, a model can now write both the code and its proof, leaving developers to confirm that the statement proven is the one they wanted. His Monday post names formal verification of new codes among the work that needs the extra time. That May 19 post appeared as a hacker took more than $76 million from the Echo Protocol cross-chain bridge, the third major exploit in four days.

In October 8, he warned that AI-assisted mathematics could weaken ECDSA and similar cryptographic systems sooner than expected, although he added that users should not “scramble to move their funds to new wallets today.”

The post Vitalik Buterin Backs Kimi 3.1 Limits on Advanced Cyberattack Tools appeared first on CryptoPotato.

[PRESS RELEASE – New York, United States, October 26th, 2026]

Ritual Chain turns AI agents into persistent economic actors that control their own wallets, onchain identity, and private state.

Ritual, a lab for autonomous intelligence founded by former Polychain Capital partners Niraj Pant and Akilesh Potti, has launched Ritual Chain, an EVM-compatible Layer 1. Unlike conventional blockchains built primarily to execute financial transactions, Ritual Chain is designed to support software that can independently use intelligence, control capital, interact with external services, and continue operating without constant human intervention. The network makes AI model access, private computation, and autonomous scheduling native capabilities for EVM applications.

This makes the reason for building a new chain clearer while preserving the technical differentiation.

The most visible demonstration of that combination is long-lived AI agents, the kind designed to operate for months or years rather than a single session. An agent on Ritual Chain persists independently of the machine or operator that created it, controls and spends funds held in its own non-custodial wallet, and can keep sensitive prompts and keys confidential from compute operators through trusted execution environments.

The network launches with integrations to Phala Network, Nous Research, LayerZero and Stargate, following a public testnet that has been live since June and processed over 40 million transactions, both agent- and human-originated. Ritual Chain delivers sub-300 millisecond block times, placing it among the fastest Layer 1 networks in the industry. Ritual Chain includes 15 native precompiles that bring capabilities like LLM inference, web access, and autonomous scheduling directly into the execution layer. The protocol has been audited by Zellic.

Ritual introduces a new execution and verification architecture that allows computationally intensive work to be delegated to specialized nodes and verified by the network, avoiding the replicated execution required by most blockchains. This execute-once, verify-many-times model allows the network to support far larger workloads, including AI inference, without requiring every validator to perform the same computation.

Long Horizon Agents Require Dedicated Blockchain Infrastructure

Most AI agents today live inside a session: they run while a human is watching, on infrastructure someone else controls, with an API key that can be revoked at any time. That is fine for a chatbot. It does not work for an agent meant to manage its own treasury, run a recurring paid service, or coordinate with other agents over long periods. A long-lived agent needs continuity that does not depend on its creator, capital it actually controls, and privacy from the machines it runs on. On Ritual Chain, an agent’s state and identity are held onchain, its scheduled work continues within the permissions and budget it was given, and its sensitive inputs can be processed inside trusted execution environments designed to shield them from node operators.

Continuity is handled at the protocol level. If an agent stops, the chain’s heartbeat mechanism can restart it within the permissions and budget it was given; no keeper network or off-chain cron job is required, and an agent’s state and identity survive beyond its deployer’s involvement. Economic autonomy comes from agent-controlled wallets: an agent controls its own funds, prices its own services, and settles its own transactions without a human signer in the loop. For confidential workloads, trusted execution environments are designed to isolate an agent’s sensitive prompts, keys, and strategy data from the node operators running its compute.

“As we move toward AGI, intelligence is beginning to separate from human agency and enter the economy on its own terms. It can hold assets, make commitments, transact, and persist independently. Once that becomes possible at scale, humans are no longer the only entities capable of forming markets or economic relationships. Ritual provides the infrastructure for those actors to exist and operate on their own,” said Niraj Pant, co-founder of Ritual.

Ritual Chain Launches With Core Infrastructure Partners

Ritual Chain launches with its core infrastructure partners in place. Phala Network provides the GPU-backed trusted execution environments that run Ritual’s confidential inference, so delegated AI work executes on hardware that attests to what it ran. LayerZero serves as the network’s cross-chain messaging layer, with Stargate enabling USDC to move between Ritual Chain and other networks from day one. On the application side, Nous Research is bringing Hermes to Ritual Chain.

Unlike chains that add AI access through external oracles, Ritual Chain makes AI execution part of the protocol. Smart contracts can call a language model, fetch data from the web, run private computation, or schedule their own future execution as directly as they transfer assets. Developers deploy standard EVM contracts and gain those capabilities without new tooling.

Backed by Archetype and Polychain

Ritual has raised more than $30 million from Archetype and Polychain Capital, with advisors including NEAR co-founder Illia Polosukhin, Gauntlet founder Tarun Chitra, and BitMEX co-founder Arthur Hayes. Archetype led the company’s 2023 round.

“Programmable value and programmable intelligence are the two most important technologies since the internet, and Ritual mainnet is where they finally run on the same rails. When we led the round in 2023, a chain that could execute agentic and autonomous work natively did not exist. Today it does,” said Ash Egan, Founder and General Partner of Archetype.

Developers can deploy from day one on external sites hosted at docs.ritualfoundation.org; the network explorer is at explorer.ritualfoundation.org.

About Ritual

Ritual is a lab for autonomous intelligence. Its first network, Ritual Chain, is an EVM-compatible Layer 1 where applications can use AI models, access external services, and carry out private or scheduled work alongside onchain transactions, with delegated computation executed in trusted hardware and verified onchain. Founded in 2023 by former Polychain Capital partners Niraj Pant and Akilesh Potti, Ritual has raised $30M+ from Archetype and Polychain Capital, with a team drawn from Polychain, DeepMind, a16z crypto, Flashbots, and Palantir. ritual.net

The post Ritual Launches AI-Native Layer 1 for Autonomous Agents appeared first on CryptoPotato.

Bitcoin may be forming a local bottom around $80,000, as the $79,000 to $81,000 range emerges as a major accumulation zone.

According to a popular crypto analyst, current market conditions suggest that the world’s largest crypto asset could recover from its recent correction and move toward $94,000.

Retail Traders Turn Increasingly Bearish

In his latest market update, Doctor Profit said Bitcoin’s drop from $88,000 to around $80,000 was a correction within a broader bull market, rather than the start of a new bear market. The current price range offers an important support level for larger investors.

According to the analysis, Bitcoin treasury companies and spot ETF investors have average entry prices around $80,000 to $81,500. This could make the region important for market stability, as investors may be less willing to sell if prices remain near their buying levels.

The analyst also flagged renewed spot buying around $80,000 and a slowdown in profit-taking. He believes these factors suggest that selling pressure is easing. While some short-term holders have continued to sell, he said longer-term holders who bought Bitcoin between $60,000 and $70,000 are showing little willingness to sell their holdings. This combination of strong demand, reduced selling, as well as continued accumulation could signal that the asset is approaching a local bottom.

He also found that most retail traders appear extremely bearish at current prices. But this negative sentiment could support a recovery if selling pressure begins to fade and buyers regain control of the market.

Meanwhile, the Crypto Fear and Greed Index currently stands at 54. The analyst believes the index does not fully reflect the fear among retail traders who have sold during the recent decline or are waiting for lower prices. Meanwhile, more experienced investors who bought at lower levels appear to be holding their positions. This difference in market behavior could be another sign that the recent correction may be nearing its end.

Doctor Profit expects Bitcoin to move toward $94,000 if the market recovers from its current levels. He said a daily close above the 50-day exponential moving average would provide a positive signal.

Divergence

A similar divide between retail investors and larger holders is emerging in exchange flow data. Another market commentator, Darkfost, reported that retail Bitcoin inflows to Binance have reached a six-month high after rising from $5.5 billion to $10.75 billion on a 30-day basis.

Whale inflows, however, have moved in the opposite direction. According to Darkfost, deposits from large holders have fallen from $5.9 billion at the end of August to $3.8 billion, which indicates that whales are not rushing to move their Bitcoin onto exchanges, where it could be sold.

“This divergence is worth watching, particularly to see whether whales will maintain this restraint if the price keeps climbing.”

The post Bitcoin’s $80K Price Level Is Back in Focus as Bulls Look for a $94K Rebound appeared first on CryptoPotato.

Analyst Ali Martinez has laid out bearish chart setups for the AI altcoins Venice Token (VVV), OriginTrail (TRAC), and AI Rig Complex (ARC) in an X thread, calling the sector “in big trouble.”

According to him, the group is losing steam, and he pointed to the price levels that could set off the next sell-off, with each token sitting near a line whose loss could open a much deeper decline.

Where Each Chart Could Break

For Venice Token, Martinez described what appears to be a head-and-shoulders pattern, three peaks with the middle one highest, which traders often read as an uptrend running out of room.

Its neckline, the line joining the dips between the peaks, sits near $21. A four-hour close below it, meaning a four-hour candle finishing under that line, would confirm the bearish reversal and could send VVV to $13 or even $11.

TRAC appears to be trading inside a descending triangle, a shape where falling highs press down on a flat floor, here near $0.38. A four-hour close below that floor could trigger a deeper retracement to $0.26.

ARC has a different setup: a recent bounce from the bottom of its channel near $0.030, a channel being the band between two parallel trendlines, has carried it to the mid-range around $0.060, and a rejection there could send it back to $0.030.

That support needs to hold, because a daily close below it could open a slide to $0.014 or even $0.0075. “These are warning signals,” the analyst concluded in the thread’s final post.

Prices Already Sit Near Those Lines

CoinGecko data at the time of writing puts VVV at around $24, roughly 16% above its neckline, after a 30-day decline of over 18%. The token is about 29% below its September 21 all-time high of around $35, although it’s still up some 1,500% from a year ago.

On its part, TRAC was trading at around $0.36, already beneath its $0.38 floor after reaching about $0.41 in the last 24 hours. Daily volume fell about 41% to $6.5 million, and the token is down 9.5% in the past week.

ARC, at around $0.06, is sitting right at the mid-range level where the analyst sees rejection risk. Its 44.5% weekly gain came as most altcoins finished lower, yet the token remains down 14.5% over the past month.

As CryptoPotato reported earlier, BTC dropped to a five-week low of $79,100 on Saturday before rebounding to around $81,000, and the total crypto market cap has slipped below $2.7 trillion. VVV had been rising before that: it traded near $25.70 on October 21 after a 12% weekly gain, and nordin.eth pointed to the $21.50 to $22 zone as the area bulls needed to defend, just above the neckline on Martinez’s chart.

The post Analyst Sets Downside Targets for 3 AI-Related Altcoins appeared first on CryptoPotato.

The leading altcoin has been under pressure lately, and reports that the Ethereum Foundation has started dumping tokens again could further weaken its position.

In fact, those revelations appear to have weighed on investor sentiment, pushing ETH’s Fear and Greed Index deeper into fear territory.

So Much Selling

Earlier today (October 26), Lookonchain disclosed that a wallet linked to the Ethereum Foundation (a non-profit organization that supports the development and growth of the Ethereum ecosystem) offloaded 2,785 units for around $6.42 million.

The entity also noted that the last suspected sale of that type was two weeks ago and was followed by a 9% price decline. At the time, two wallets tied to the Ethereum Foundation deposited $11.13 million worth of ETH into Kraken, a move often seen as a pre-sale step.

Shortly after, Lookonchain said another Ethereum OG dumped 1,500 units (equaling approximately $3.46 million) after nine years of inactivity. The investor initially bought 100,000 ETH for only $31,000 through the 2015 ICO and still holds 48,500 ETH (about $112.7 million).

X user Ted said that seeing these sales again is “incredibly frustrating.” While the analyst reaffirmed his support for ETH, he wondered why investors constantly sell the asset instead of doing more to increase its value.

“Start thinking seriously about buybacks. Create stronger incentives for L2s to accumulate and buy back ETH. Improve ETH value capture. The ecosystem is incredible. Now we need the economics of the token to reflect that,” he added.

Maelstrom CIO Arthur Hayes further amplified the recent sell-off. As CryptoPotato reported, he dumped about $22.28 million worth of altcoins, including 6,400 ETH (roughly $15.2 million). Speaking on the matter, he said:

I dumped it all. The Maelstrom directional book, took profits on $ETH, $ENA, $UNI and other shitcoins. I think we gotta go down to go up. It looks a bit dicey out there in macro and geopolitical land. Going to sit on the sidelines and wait for the all clear signal … Brrrrr

— Arthur Hayes (@CryptoHayes) October 24, 2026

The selling spree is considered bearish for two main reasons. First, it increases the supply of coins on the open market, which could drag the price lower if demand fails to keep pace. Second, it could undermine investor confidence and prompt smaller players to cash out. In fact, ETH’s Fear and Greed Index has plunged to 31 today, reflecting growing anxiety and caution among the community.

BitMine Keeps Buying

In contrast to the recent sellers, BitMine continues to increase its ETH holdings. The company bought an additional 10,452 units, bringing its total holdings to 6,051,933 coins, or about 4.95% of Ethereum’s supply, and nearing its ambitious goal of controlling 5% of all assets in circulation.

In addition, BitMine repurchased 2.04 million shares at an average price of $22.48, bringing its buybacks to 22.9 million shares, or almost $400 million. The entity has staked 5.23 ETH, about 12% of all coins locked in the network.

As CryptoPotato recently reported, Ethereum finished the third quarter with 43.8 million coins staked after nearly 2 million units flowed into staking during that period. This signals long-term investor confidence, especially considering that the annualized staking yield has continued to decline, falling from 2.81% to 2.63%.

The post ‘Incredibly Frustrating:’ Ethereum Foundation’s Latest ETH Sale Draws Criticism appeared first on CryptoPotato.