CCTV News reported that US President Trump said on June 13 local time that the United States and Iran plan to sign an agreement on June 14, and the Strait of Hormuz will reopen immediately after signing. Multiple mediating parties said the signing will be held online due to scheduling and logistics, while Iranian officials said the exact timing still needs confirmation.

Google has filed a lawsuit over what it says was a large-scale phishing and SMS scam operation that used the company's Gemini AI to target hundreds of thousands of U.S. victims, including cryptocurrency users. In a complaint filed June 12, Google names an alleged China-based cybercrime network known as Outsider Enterprise. The company claims the group used Gemini to help automate fraud at scale, generating code and templates for fake websites designed to mimic legitimate telecom portals and other services. Federal authorities describe the operation as sprawling. The FBI says the network stood up more than 8,000 phishing sites across dozens of countries. Google also says it received about 55,000 reports of suspicious messages in Google Messages during the two-week period ending June 1, many of which it links to Outsider Enterprise. Court filings estimate the network stole roughly 3.87 million credit card numbers and contributed to about $1.9 billion in losses since July 2023. Google's suit argues that crypto holders have become a prime target, with scammers aiming at cryptocurrency wallets and exchange logins. The filing points to limited recovery options for many digital-asset victims compared with traditional banking customers. The broader FBI numbers underscore the scale of crypto-related fraud. In 2025, the bureau logged 1,008,597 internet crime complaints. Crypto-related reports totaled 181,565, with $11 billion in losses, the largest category by dollar amount. AI-enabled scams are increasingly prominent. The FBI's Internet Crime Complaint Center (IC3) has, for the first time, included a dedicated section on AI-driven fraud. In 2025, those cases generated 22,364 complaints and nearly $893 million in reported losses. Google characterizes the lawsuit as an effort to "permanently dismantle" a group it says used AI tools, including Gemini, to run fraudulent text-message campaigns. The case comes as researchers and regulators warn that even leading AI models can be repurposed for harmful activity, a concern that has intensified as major tech firms embed generative AI into consumer products. Law enforcement has also stepped up efforts. The FBI's Operation Level Up, launched in 2024, has notified more than 8,000 cryptocurrency fraud victims and helped avert more than $500 million in potential losses. The lawsuit is a notable civil attempt to hold alleged developers and core operators within a cybercrime network accountable for using generative AI to scale financial fraud. The dispute highlights how advanced AI can accelerate traditional phishing tactics and sharpens the focus on the intersection of AI, organized cybercrime, and vulnerabilities across crypto ecosystems. If successful, the case could serve as an early test of how courts and platforms respond when generative AI becomes a production tool for large-scale financial fraud.

Data from MyToken show that, as of June 14 (UTC+8), liquidations across the cryptocurrency derivatives market totaled $142 million over the past 24 hours. BTC accounted for $35.872 million, ETH for $24.7806 million, SOL for $3.5222 million and XRP for $544,900. (Source: MyToken)

Anthropic said a U.S. government emergency export-control directive compelled the company to cut off access to two flagship models—Claude Fable 5 and Claude Mythos 5—for foreign nationals, including foreign-national employees. The company said it received the order at 5:21 p.m. ET on June 12 and executed a worldwide shutdown of the two models to meet compliance requirements, describing the action as driven by national-security concerns. Anthropic emphasized that other products remain unaffected, including Claude Opus 4.8, which the company said continues operating normally. The news quickly spilled into crypto markets through pre-IPO-linked instruments that some crypto-native traders use to gain exposure to private technology names. On Hyperliquid, the Anthropic perpetual contract reportedly fell about 3.7% to roughly $1,627, down from post-launch highs above $1,800, with open interest around $8.6 million. The move underscored how AI regulatory headlines can transmit into crypto price action in near real time. Anthropic said the directive followed reports of a "non‑universal jailbreak" affecting Fable 5. According to the company, the government provided only verbal evidence tied to a narrow prompt that asked the model to review a specific codebase and identify software flaws. Anthropic argued the flaws were minor, previously known, and discoverable using public models without any bypass, saying the situation did not warrant a recall-style shutdown of a commercial system. Anthropic also warned that applying the same restriction threshold broadly could slow or effectively halt deployments of frontier AI models across the industry, turning an expansive standard into a de facto barrier to releasing advanced systems. For crypto markets, the episode highlights two dynamics: tokenized or pre-IPO derivatives linked to private tech companies are increasingly used as high-velocity sentiment gauges, and they can reprice regulatory and national-security uncertainty faster than supporting evidence becomes public—amplifying volatility when information is incomplete. The government's technical report has not been released. Anthropic said it has only received verbal evidence so far, leaving markets to balance the company's account against an opaque national-security process. Uncertainty remains elevated around Anthropic's model access, regulatory exposure, and risks to frontier-model deployment. Source: Anthropic's official statement.

ME News reported that on June 14 (UTC+8), prediction market platform Polymarket issued a "resolution clarification" that overturned what had appeared to be a settled market result. The change nullified a $35,000 prediction placed by a 20-year-old student and eliminated roughly $3.8 million in open positions across 1,838 accounts. The platform's terms of service include a clarification provision that allows after-the-fact interpretive adjustments to market resolutions, which can alter final payouts. Traders criticized the move as a retroactive reversal that weakens confidence in the rules and has fueled controversy across both the Polymarket and Kalshi communities. User accounts trace the episode to a case made public on June 13: a market outcome initially appeared finalized, then was later reversed after a reinterpretation of the rules. Analysts say the episode highlights "resolution clarification risk"—a non-hedgeable tail-risk event. If similar actions become common, high-risk liquidity could migrate from current platforms to CFTC-regulated venues or exchanges that offer formal arbitration. The dispute is also being viewed alongside a recent string of controversies—including disagreements over UMA oracle settlements tied to Bitcoin markets—that continue to test participants' trust in the finality of prediction market outcomes. (Source: ODAILY)

Google filed a lawsuit on June 12 against a suspected China-based cybercrime network it identifies as Outsider Enterprise, alleging the group used the company's Gemini AI to automate large-scale SMS phishing campaigns that impersonated legitimate brands and harvested financial credentials from hundreds of thousands of victims. Court documents say the defendants relied on Gemini to generate code and website templates that produced convincing phishing pages mimicking telecom portals and other services. The FBI alleges the operation created more than 8,000 phishing sites across dozens of countries. Google said it received about 55,000 reports of suspicious messages in Google Messages during the two-week period ending June 1, many of which it believes were connected to Outsider Enterprise. Filings estimate the network stole roughly 3.87 million credit card numbers and drove about $1.9 billion in losses since July 2023. Crypto users were also in the crosshairs. In addition to payment cards and account logins, the phishing infrastructure targeted cryptocurrency wallets and exchange credentials. The focus on digital-asset holders reflects a broader shift: crypto victims often have fewer recovery options than traditional banking customers, reinforcing the need for stronger custody and access controls such as hardware wallets and robust multifactor authentication. The suit lands as AI-enabled fraud accelerates across the U.S. The FBI reported 1,008,597 internet crime complaints in 2025, with crypto-related cases the largest category at 181,565 reports and about $11 billion in losses. For the first time, the FBI's Internet Crime Complaint Center (IC3) added a dedicated section on AI scams in 2025, logging 22,364 complaints that cost Americans nearly $893 million. Separately, the FBI's Operation Level Up, launched in 2024, has identified more than 8,000 crypto fraud victims and helped avert over $500 million in potential losses. Google said the goal of the case is to "permanently dismantle" the operation's core software developers. A successful outcome could set a notable legal precedent for holding operators and developers responsible for weaponizing AI tools. The episode also reinforces industry warnings that advanced AI models can be repurposed for fraud, an issue gaining urgency as major consumer tech companies, including Apple, integrate AI features into widely used products. For crypto readers, the takeaway is straightforward: generative AI is increasing both the scale and sophistication of phishing, and digital-asset holders remain prime targets. Pressure is likely to rise on platforms, exchanges, and AI providers to strengthen defenses, while regulators and law enforcement pursue more cross-border takedowns. Basic hygiene still matters: store assets in hardware wallets, use strong MFA, and treat unsolicited SMS links with skepticism.

A nationwide survey of 51,993 Americans highlights a conflicted view of artificial intelligence: 64% cite job losses as their biggest fear, while the leading hope is AI's potential to cure diseases such as cancer. Employment anxiety topped concerns in every state, ranging from 71% in Iowa to 57% in Mississippi, and appeared across party lines. Daily AI users were markedly less worried than nonusers (54% vs. 70%). Other major concerns included cognitive dependency (56%) and misinformation (52%). Only 15% said they trust AI companies to manage the technology responsibly, the lowest trust score among all institutions tested. The public unease tracks recent labor data. AI was linked to 38,579 US job cuts in May, about 40% of the month's total. Employers have attributed 87,714 reductions to automation so far in 2026, already above the 54,836 tied to the technology across all of 2025. In Washington, Senators Elizabeth Warren and Bernie Sanders have urged Congress to strengthen worker protections. Not all industry leaders agree: Amazon founder Jeff Bezos has dismissed the displacement thesis and instead predicts labor scarcity, even as his AI venture Prometheus raised $12 billion at a $41 billion valuation. Separately, the US government issued an emergency export-control directive on Friday ordering Anthropic to immediately suspend access to its two most advanced models—Claude Fable 5 and Claude Mythos 5—for all foreign nationals, including Anthropic's own employees, citing national-security concerns tied to a potential jailbreak. Anthropic said the order's scope was so broad it disabled the models for its entire customer base to ensure compliance. Both models had been released only days earlier. Mythos 5, which was described as having fewer guardrails and a strong ability to uncover cybersecurity exploits, had been limited to a small set of select partners. Anthropic said it complied but disputed the government's conclusions, calling the action an overreach. The company said it received only verbal evidence of a narrow, nonuniversal vulnerability—essentially prompting the model to review a codebase and fix software flaws—and argued similar capabilities are already widely available from rival systems, including OpenAI's GPT5.5. Anthropic warned that applying this standard industrywide would effectively stop new frontier-model deployments. Access to its other models is unchanged, and the company said it is working to restore the suspended ones. Markets were also absorbed by what was billed as the largest public offering ever. SpaceX priced its IPO at $135 per share on June 11, raising $75 billion. Shares opened on Nasdaq at $150 on Friday morning and climbed to $164. The listing quickly spilled into crypto markets: retail traders sought exposure not only via Nasdaq shares but also through a redeemable token on Solana, tracker certificates on multiple exchanges, a wallet subscription campaign, and perpetual futures—a rare convergence of traditional and onchain rails around a single equity. That crossover put a spotlight on a central fault line in tokenized stocks: the same ticker can represent very different legal claims. Onchain, Backpack Securities' SPCX token is backed 1:1 by a real share held in custody and is redeemable for the underlying equity through regulated ACATS/DTCC processes. By contrast, xStocks tracker certificates on certain exchanges offer no shareholder or voting rights, and their collateral may not always be the underlying shares. On the DEX side, Hyperliquid's cash-settled perpetual futures provide no claim on the company at all. Together, the developments underscore how quickly AI and tokenization are advancing relative to frameworks governing ownership, control, and accountability. A key primary-source detail comes from SpaceX's own listing disclosures. According to the broker-dealer's official investor disclosure, the SPCX token on Solana was launched to coincide precisely with the Nasdaq debut, marking the first time a newly listed equity had a simultaneous onchain market from day one, with shares priced at $135 and trading opening at $150 Friday morning. COINOTAG said this redemption-backed structure should be treated as a benchmark distinguishing true equity rights from simple price exposure.

ME News reported that on June 14 (UTC+8), SPCXUSDC led all assets by 24-hour trading volume, posting $988.8 million in notional turnover. BRENTOILUSDC had the highest funding rate among tracked markets, reaching 0.0033 over an eight-hour window. Funding rates for all five spot-linked assets stayed positive, suggesting derivatives positioning remained broadly balanced with no signs of an outsized long bias. The five most active assets were SPCX, XYZ100, CL, BRENTOIL and SP500. (Source: D Pro)

Crypto markets turned choppy again, setting off another wave of forced unwinds. Over the past 24 hours, total liquidations reached about $186 million as sharp price moves squeezed leveraged traders on both sides. Longs took the larger hit at $102.8 million, while shorts accounted for $83.2 million, reflecting a market lacking clear direction. Source: CoinGlass Bitcoin [BTC] led the liquidation totals with $34.97 million, followed by Ethereum [ETH] at $24.65 million. The near-even split between long and short losses points to uncertainty rather than strong conviction, keeping markets exposed to additional swings if macro conditions stay tight. Whale selling offsets ETF inflows Derivatives positioning was whipsawed by volatility, while spot flows told a more mixed story. Over the past month, whales distributed more than 70,000 BTC, adding supply as Bitcoin traded below prior highs. Source: X The selling suggests some large holders remain cautious amid shifting macro expectations and uncertain liquidity. Even so, demand has not disappeared. U.S. Spot Bitcoin ETFs posted $85.85 million in net inflows on June 12, signaling continued institutional allocation despite recent weakness. This push-pull helps explain why Bitcoin has softened without sliding into broad capitulation: persistent distribution is being met by steady ETF demand, leaving neither side firmly in control. Ethereum exchange supply tightens Ethereum's supply dynamics moved the other way. Exchange balances hovered around 15.5 million ETH early in the period, then fell toward 15.0 million ETH. The drop aligned with nearly 500,000 ETH—worth roughly $800 million—leaving trading venues within a week. That reduces the amount of ETH readily available for sale, even as the broader backdrop remains fragile. Source: X The divergence stands out given that Bitcoin whales distributed more than 70,000 BTC over the same stretch. While additional BTC entered circulation, Ethereum's exchange-held supply kept shrinking, suggesting selling pressure is more evident in Bitcoin, while Ethereum's market structure continues to tighten beneath the surface. Final Summary Recent liquidations in Bitcoin and Ethereum underscore a market still heavily influenced by leverage and macro uncertainty. Bitcoin is absorbing fresh supply from whale selling, while Ethereum continues to draw down exchange inventories.