Invest ideasNews
28 May 2026, 01:00
Bitcoin Drop Linked To Hidden $1.3 Billion ETF Trade, Analyst Reveals
Eight straight days. That is how long US spot Bitcoin ETFs have been bleeding money, with more than $2 billion in net outflows recorded since May 14 — and Tuesday’s session added another ugly chapter to that streak. Related Reading: HYPE Price Breakout Ignites Rally Talk Toward $170 Target A Sell Order Like No Other A single trader sold over 29 million shares of BlackRock’s iShares Bitcoin Trust ETF on Tuesday through a dark pool, a private trading platform used by institutions to quietly execute large orders away from public markets. The transaction, valued at $1.3 billion, was executed at $43.16 per share at 2:30 pm UTC. Alex Thorn, head of firmwide research at Galaxy Digital, said it was the biggest dark pool trade in the fund he had ever seen. Bloomberg ETF analyst Eric Balchunas added more weight to that claim, pointing out that the sell order was more than 22 times larger than the second-biggest IBIT sell order recorded on the same day. The identity of the trader has not been disclosed. massive $1.289 billion IBIT block sale by unknown party through dark pool at 10:30am today, biggest such trade i’ve ever seen pic.twitter.com/9qGDqkfCbu — Alex Thorn (@intangiblecoins) May 26, 2026 Bitcoin Slid Within Minutes Price data from TradingView shows Bitcoin fell 1.45% — from $77,870 to $76,721 — within 10 minutes of the trade being executed. The drop did not stop there. Bitcoin continued sliding and hit a 24-hour low of $75,600 roughly 12 hours later, marking a 2.5% loss for the day. Tuesday’s total outflow from US spot Bitcoin ETFs came in at $333 million, with IBIT alone accounting for over $192 million of that figure. That brings the cumulative outflow since May 14 past the $2 billion mark. Institutions Pulling Back The sell-off fits a broader pattern of institutional retreat. Jane Street cut its Bitcoin ETF holdings by around 70% in the first quarter, while Goldman Sachs trimmed its position by 10%. Bitcoin has historically traded outside the orbit of traditional financial markets, but the rise of US-based Bitcoin ETFs has pulled institutional investors in — and now some are heading for the exit. Related Reading: When Bitcoin Gets Ignored, It Tends To Rally The Hardest, Analyst Says Fresh capital entering the market has not been enough to offset the pace of withdrawals, according to reports. Whether Tuesday’s massive dark pool transaction signals a shift in strategy by a major holder, or simply a one-time portfolio move, remains unknown. Featured image from Unsplash, chart from TradingView
28 May 2026, 01:00
Whale Opens $25.5 Million Ethereum Short on Hyperliquid as Singer’s Long Position Liquidated
BitcoinWorld Whale Opens $25.5 Million Ethereum Short on Hyperliquid as Singer’s Long Position Liquidated A prominent anonymous trader known as Evaded has opened a substantial short position on Ethereum, worth $25.49 million, on the decentralized exchange Hyperliquid. The move, executed with 25x leverage, involves 12,600 ETH and was first flagged by on-chain analytics platform Onchain Lens. Details of the Short Position The short position, opened on March 20, 2025, adds to Evaded’s existing bearish bets. The trader is also maintaining a 30x leveraged short position on Bitcoin valued at $71.5 million. According to available data, the Bitcoin short is currently showing unrealized profits exceeding $1.6 million, suggesting a degree of market timing success. The combined short exposure across both assets now totals nearly $97 million. Liquidation of a Celebrity Long Position In a contrasting development, a portion of a 25x leveraged long position on Ethereum held by prominent Taiwanese singer and crypto investor Jeffrey Huang has reportedly been liquidated. Huang, known for his active trading on Hyperliquid, had previously disclosed large long positions. The liquidation event underscores the risks inherent in high-leverage trading, where even minor price movements can trigger forced closures. Why This Matters for the Market These movements highlight the growing influence of decentralized exchanges like Hyperliquid, which allow for high-leverage trading with minimal barriers. The actions of a single whale can create ripple effects, particularly in thinner order books. For retail traders, these events serve as a reminder of the volatility and risks associated with leveraged positions. The divergence between Evaded’s bearish stance and Huang’s liquidated long position also reflects the uncertainty currently surrounding Ethereum’s price direction. Conclusion The opening of a $25.5 million Ethereum short by a known whale, alongside the liquidation of a celebrity trader’s long position, provides a snapshot of the high-stakes environment on Hyperliquid. While the whale’s existing Bitcoin short is profitable, the new Ethereum bet introduces additional risk. These events are a clear signal for traders to monitor leverage levels and market sentiment closely. FAQs Q1: Who is Evaded? Evaded is an anonymous trader known for large, high-leverage positions on Hyperliquid. Their identity is not publicly known. Q2: What is Hyperliquid? Hyperliquid is a decentralized exchange (DEX) that offers perpetual futures trading with high leverage, often up to 30x or more. Q3: What does liquidation mean in crypto trading? Liquidation occurs when a trader’s position is forcibly closed by the exchange because the margin (collateral) has fallen below the required level, usually due to adverse price movements. This post Whale Opens $25.5 Million Ethereum Short on Hyperliquid as Singer’s Long Position Liquidated first appeared on BitcoinWorld .
28 May 2026, 00:50
BTC Spot CVD Analysis: Volume Heatmap Signals Key Levels on May 28
BitcoinWorld BTC Spot CVD Analysis: Volume Heatmap Signals Key Levels on May 28 An analysis of the BTC/USDT spot order book using the Spot Cumulative Volume Delta (CVD) chart as of 12:00 a.m. UTC on May 28 reveals nuanced trading activity that may help traders identify key support and resistance levels. The chart combines a Volume Heatmap in the upper section with a CVD indicator below, offering a layered view of market structure. Understanding the Volume Heatmap The Volume Heatmap tracks the volume of executed trades at specific price levels. When the price lingers in a certain range or moves significantly, the background color on the heatmap brightens. These brighter areas often act as visual markers for potential support and resistance zones, as they represent price levels where a high volume of trades has occurred. Traders frequently monitor these zones for potential price reactions. Decoding the Cumulative Volume Delta (CVD) The CVD indicator at the bottom of the chart categorizes buy and sell orders by trade size. As buy orders for a specific category increase, the corresponding colored line rises. For instance, the yellow line tracks orders between $100 and $1,000, while the brown line represents large orders between $1 million and $10 million. This segmentation allows traders to see which size cohorts are driving market momentum, offering insight into whether retail or institutional activity is more influential at a given moment. Implications for Traders By observing which CVD lines are rising or falling, traders can gauge the aggressiveness of buying or selling pressure from different market participants. A rising brown line (large orders) alongside a flat or declining yellow line (small orders) may indicate institutional accumulation, which could signal a stronger trend. Conversely, a surge in small orders without large-order confirmation might suggest retail-driven noise rather than a sustainable move. Conclusion The BTC spot CVD chart for May 28 provides a granular view of order flow dynamics, helping traders identify key levels and the types of participants influencing price action. While no single indicator is predictive, the combination of volume heatmap and CVD analysis offers a factual framework for assessing market structure. Traders should use this data alongside other tools and risk management practices. FAQs Q1: What is the Spot Cumulative Volume Delta (CVD)? The CVD is an indicator that tracks the net difference between buy and sell orders in the spot market, categorized by trade size. It helps traders understand the strength and direction of order flow. Q2: How does the Volume Heatmap help identify support and resistance? The heatmap highlights price levels where a high volume of trades has occurred. These levels often act as support (where price stops falling) or resistance (where price stops rising) because of the concentration of past trading activity. Q3: What does a rising CVD line for large orders indicate? A rising CVD line for large orders (e.g., $1 million to $10 million) suggests that institutional or high-net-worth traders are actively buying, which may signal a stronger, more sustained price move compared to retail-driven activity. This post BTC Spot CVD Analysis: Volume Heatmap Signals Key Levels on May 28 first appeared on BitcoinWorld .
28 May 2026, 00:45
Dormant Whale Moves $7 Million in Ethereum to Kraken After Two Years
BitcoinWorld Dormant Whale Moves $7 Million in Ethereum to Kraken After Two Years A previously dormant cryptocurrency whale has reactivated, depositing 3,466 ETH — worth approximately $7 million — to the Kraken exchange roughly two hours ago, according to on-chain analytics platform Onchain Lens. The wallet address, which begins with 0x9295, had shown no significant activity for the past two years. What the On-Chain Data Reveals Deposits of large amounts of cryptocurrency to centralized exchanges are widely interpreted by market analysts as a preparatory step toward selling. The sudden movement of funds from a long-dormant wallet often draws attention, as it can signal a shift in sentiment from a major holder. While the identity of the wallet owner remains unknown, the transaction was recorded on the Ethereum blockchain and verified by multiple block explorers. Context and Market Implications This whale movement comes at a time when the broader cryptocurrency market is experiencing mixed signals. Ethereum’s price has shown volatility in recent weeks, influenced by macroeconomic factors and regulatory developments. Large transactions, especially those moving funds to exchanges, can contribute to short-term selling pressure, though the overall impact depends on whether the assets are actually liquidated. Why This Matters for Traders and Investors For retail investors and market observers, tracking whale activity provides valuable insight into the behavior of large capital holders. While a single deposit does not guarantee an immediate sell-off, it is a data point that, when combined with other on-chain metrics, helps form a more complete picture of market sentiment. The reactivation of a wallet after two years of dormancy adds an extra layer of significance, as it suggests a deliberate decision by the holder to re-engage with the market. Conclusion The $7 million ETH deposit to Kraken by a previously inactive whale is a notable on-chain event that underscores the importance of monitoring large wallet movements. While the ultimate intention of the holder remains unclear, the transaction serves as a reminder of the transparency and analytical value inherent in public blockchain data. FAQs Q1: What is a cryptocurrency whale? A whale is an individual or entity that holds a large amount of a particular cryptocurrency, enough to potentially influence market prices through their trading activity. Q2: Why is a deposit to an exchange considered a potential sell signal? When a whale moves funds from a private wallet to an exchange, it is often interpreted as a preparatory step for selling, as exchanges are the primary platforms for converting crypto into fiat currency or other assets. Q3: How reliable is on-chain data for predicting market movements? On-chain data provides transparent and verifiable transaction records, but it should not be used in isolation. It is most valuable when combined with other market indicators, such as trading volume, order book depth, and broader economic trends. This post Dormant Whale Moves $7 Million in Ethereum to Kraken After Two Years first appeared on BitcoinWorld .
28 May 2026, 00:44
Google engineer allegedly had a cheat code for Polymarket and turned it into $1.2M
Federal prosecutors have charged a Google software engineer in a case involving alleged insider-style trading on the blockchain prediction platform Polymarket. According to federal prosecutors, 36-year-old Michele Spagnuolo knew the results months before anyone else. He was arrested on May 27, 2026. The US Attorney’s Office for the Southern District of New York released a criminal complaint that charges him with commodities fraud, wire fraud, and money laundering. Google holds an annual “Year in Search” campaign, where it reveals the most-searched people, events, and topics, and people usually bet on the results before they come out. What is Polymarket, and how does someone make money on it? Polymarket is an online prediction market where people bet on the occurrence of real-world events . It uses a simple yes-or-no system in which each “YES” and each “NO” is priced between 0 and 1 dollar. The price changes depending on what other traders think the odds are. For example, if there is a 90% chance that Donald Trump will NOT be the top-searched person, then the “NO” share for Trump trades around 90 cents. That means if Trump is indeed not the top result, and you had bought that share, you would make 10 cents per share, as each share pays out a full dollar. It is a guessing game for many traders because the system relies on assumptions, but Spagnuolo allegedly already knew the answer. What is Google’s Year in Search, and why are the results hidden? The “Year in Search” is a list of the people, events, topics, and questions that trended most on Google’s search engine during that year. The company has released the list since at least the early 2000s. According to the criminal complaint filed by FBI Special Agent Brandon Racz, the campaign has many benefits for Google. It drives millions of people to Google’s platforms and generates significant media coverage for the organization. It also reinforces Google’s status as the “authoritative barometer of public interest and cultural trends,” and gives Google a high-profile showcase to demonstrate its reach to advertisers. Google keeps the results a secret, even to most of its employees. If the results leaked early, the media buzz would disappear, advertisers would lose interest in the launch moment, and the entire marketing campaign would be undercut. Google treats this information as highly confidential and restricts the data to a small number of employees. Spagnuolo has access to the data. Which bets did Spagnuolo make? Polymarket opened two markets for the 2025 Year in Search on October 14 and 20. The first market listed about 24 people and asked which of them would be the most-searched person on Google for 2025. The second market asked whether those same people would be in the top five. The payout depended on the results that Google would publish on the Year in Search website. Spagnuolo allegedly accessed Google’s internal Year in Search tool and used an anonymous account named AlphaRaccoon to place a $403 on Kendrick Lamar as the top-searched person. The odds for that were 3%, and Google’s internal Year in Search tool already has Lamar’s name. He also bet $10,807 that Pope Leo XIV would NOT be number one, just as the list indicated. The market gave Pope Leo XIV a 50/50 chance. Spagnuolo checked the internal tool again on November 27 and saw that a musician named d4vd had replaced Kendrick Lamar as the number-one trending person for 2025. He AlphaRaccoon bet $381.12 that d4vd would be in Google’s top 5, with a market outcome of only 18% because most traders had no idea who d4vd was in the first place. The Italian also bet $5 bet that d4vd would be the single top-searched person. The market assigned a near-zero probability to that outcome, so it was basically free money for him. But the biggest bets were $937,688 that Bianca Censori, $613,587 that Pope Leo XIV, and $509,149 that Donald Trump would NOT be number one. He also added $171,612 that Donald Trump would NOT be in the top five. AlphaRaccoon risked about $2.75 million on about 25 outcome bets. Google published the results on December 4, and d4vd, Kendrick Lamar, Jimmy Kimmel, Tyler Robinson, and Pope Leo XIV made the Global Top 5. How did the FBI find Spagnuolo? Spagnuolo tried to hide all traces of the money by converting his winnings into different cryptocurrencies and running them through a service designed to erase transaction history on the public blockchain. However, the FBI traced each address and found that the same wallet belonging to AlphaRaccoon was responsible for these transactions. Here is what the US criminal complaint, Southern District of New York, stated in May 2026, “Unlike the counterparties to his trades, Spagnuolo knew the outcome of these wagers before the trading public did because he had accessed Google’s confidential, commercially valuable internal data.” How did Google respond? A spokesperson at Google published a statement after Spagnuolo’s arrest and said, “We’re working with law enforcement on their investigation. The employee accessed our marketing material using a tool available to all employees, but using such confidential information to place bets is a serious breach of our policies. We’ve placed the employee on leave and will take the appropriate action.” However, the statement contradicts the complaint. Google says the tool was available to all employees, but the criminal complaint states that Year in Search data is restricted “to only a limited number of employees” even within the company. This is the second major insider-trading incident connected to Polymarket in 2026, which is putting serious pressure on the market regarding this kind of abuse. If you're reading this, you’re already ahead. Stay there with our newsletter .
28 May 2026, 00:40
Why Google’s AI can’t spell its own name: The fundamental flaw in large language models
BitcoinWorld Why Google’s AI can’t spell its own name: The fundamental flaw in large language models Google’s AI Overview, the generative search feature the company has positioned as the future of its flagship product, continues to struggle with a task most humans master by age six: spelling. Users recently discovered that when asked how many ‘P’s are in ‘Google,’ the AI confidently answered ‘two.’ It also stated there is ‘exactly 1 r in the word poop’ and claimed there are two ‘d’s in ‘journalism’ while spelling it as ‘j-o-u-r-n-a-d-i-s-m.’ When asked about the number of ‘P’s in the U.S. president’s last name, it correctly identified one, but then spelled the name as ‘t-r-p-u-m.’ The tokenization problem These errors are not random glitches. They stem from a fundamental architectural limitation in the large language models (LLMs) that power AI Overview. Unlike humans, who perceive words as sequences of letters, LLMs break text into tokens. A token can be a full word, a syllable, or even a single character, depending on the model. The AI does not ‘read’ letters; it converts text into numerical representations that encode meaning and context. This token-based system is highly effective for generating coherent sentences and answering complex questions, but it has no inherent understanding of individual characters. Matthew Guzdial, an AI researcher and assistant professor at the University of Alberta, explained to Bitcoin World: ‘LLMs are based on this transformer architecture, which notably is not actually reading text. What happens when you input a prompt is that it’s translated into an encoding. When it sees the word ‘the,’ it has this one encoding of what ‘the’ means, but it does not know about ‘T,’ ‘H,’ ‘E.” A known but unresolved challenge Counting letters within words has been a well-documented weakness of LLMs for years. It has become something of a running joke in the AI community that whenever a new model is released, the first test should be asking how many ‘r’s are in ‘strawberry.’ Google acknowledged the issue in a statement to Bitcoin World, saying, ‘Counting within words has been a known challenge for LLMs, and we’re working to fix this particular issue.’ But fixing it may not be straightforward. Sheridan Feucht, a PhD student studying large language model interpretability at Northeastern University, told Bitcoin World: ‘It’s kind of hard to get around the question of what exactly a ‘word’ should be for a language model, and even if we got human experts to agree on a perfect token vocabulary, models would probably still find it useful to ‘chunk’ things even further. My guess would be that there’s no such thing as a perfect tokenizer due to this kind of fuzziness.’ Beyond spelling: Broader reliability concerns The spelling errors are the latest in a series of high-profile missteps for Google’s AI Overview. Earlier in May, users discovered that searching the word ‘disregard’ would return what appeared to be a dictionary definition, but the definition read: ‘Understood. Let me know whenever you have a new prompt or question!’ — a clear sign that the AI was surfacing a system prompt intended for internal use. Google has since patched that issue. These failures are not merely amusing. They underscore a deeper challenge for Google as it pushes generative AI to the center of its search experience. The company first introduced AI Overviews in 2024, only to have the feature famously advise users to eat rocks and put glue on their pizza, after citing satirical content from The Onion and Reddit. The current iteration, launched more broadly in early 2026, was supposed to be more refined. Yet basic spelling errors persist. What this means for users For the average person using Google Search, these errors are a reminder that AI, for all its impressive capabilities, is not infallible. The same models that can generate code, pass professional exams, and assist with creative writing can also fail at tasks a child can perform. The practical implication is clear: users cannot blindly trust AI-generated outputs without verification. This is especially important as Google integrates AI Overviews more deeply into search results, where users may be inclined to accept the information as authoritative. Conclusion Google’s spelling struggles are not a crisis for the company, but they are a useful reality check. They highlight that LLMs, despite their transformative potential, operate under fundamental constraints that researchers have not yet solved. The tokenization architecture that makes these models powerful also makes them blind to the building blocks of written language. As Google continues to double down on AI-driven search, these limitations will remain a source of both humor and caution for users and developers alike. FAQs Q1: Why can’t AI models like Google’s AI Overview spell correctly? LLMs use tokenization, breaking text into tokens (words, syllables, or characters) rather than reading individual letters. This makes them highly effective at understanding context but unable to reliably count or identify specific characters within words. Q2: Is Google working on fixing these spelling errors? Yes. Google has acknowledged the issue and stated it is working on a fix. However, researchers note that the problem is inherent to the transformer architecture, and a perfect solution may not exist. Q3: Should I trust information from Google’s AI Overview? AI Overviews can be useful for general information, but users should verify critical facts, especially those involving numbers, spelling, or specific details. The AI can produce confident but incorrect answers, as demonstrated by the spelling errors and the ‘disregard’ incident. This post Why Google’s AI can’t spell its own name: The fundamental flaw in large language models first appeared on BitcoinWorld .
