Follow TNGB
Finance Platform Hit by $290 Million Heist, China App Store Crypto Theft

KelpDAO Platform Hit by 290 Million Lazarus Group Heist
- KelpDAO lost approximately 290-293 million dollars in a LayerZero bridge exploit.
- Attackers leveraged single verifier configuration and DDoS on backup systems.
- LayerZero attributed the operation to North Korea’s TraderTraitor unit; Kelp disputed server compromise claims.
USA, Apr 20 (TNGB) – The decentralized finance platform KelpDAO reportedly lost between 290 and 293 million dollars in cryptocurrency during an attack that began Saturday afternoon. Hackers exploited the LayerZero cross-chain messaging bridge by targeting Kelp’s reliance on a single decentralized verifier network, which approved transactions without requiring multiple independent confirmations. They simultaneously launched a distributed denial-of-service assault on backup verification systems to prevent rapid response. LayerZero’s public post-mortem, published Monday, attributed the sophisticated intrusion to North Korea’s TraderTraitor subgroup operating within the Lazarus collective, citing preliminary indicators of state sponsorship. Kelp officials responded that the breach involved direct server compromise rather than solely configuration issues and noted that roughly 40 percent of LayerZero customers use similar single-verifier setups without prior warnings.
The incident represents the largest cryptocurrency theft recorded in 2026 to date and reveals how bridge architecture choices can create single points of failure.
Why This Matters: Configuration decisions in cross-chain protocols directly determine whether billions in user assets remain protected from state-sponsored theft operations.
China Apple App Store Infiltrated by Crypto Stealing Wallet Applications
- Twenty-six malicious apps impersonated legitimate wallets including MetaMask, Coinbase, Trust Wallet, and OneKey.
- The apps harvested seed phrases and drained cryptocurrency from user accounts.
- Kaspersky identified the campaign, which primarily targeted Chinese iOS users but contained no geographic code restrictions.
BEIJING, Apr 20 (TNGB) – Security researchers at Kaspersky documented 26 fraudulent cryptocurrency wallet applications that appeared in the Apple App Store serving Chinese users. The apps used typosquatting and counterfeit branding to mimic established products such as MetaMask, Coinbase, Trust Wallet, and OneKey, then prompted victims to enter recovery phrases or seed phrases under the guise of account setup or verification. Once obtained, the malware transferred assets to attacker-controlled wallets within minutes. The campaign focused on Chinese iOS users, yet the underlying code included no regional limitations, allowing potential expansion if operators altered distribution. Apple removed the applications after notification, though the precise number of affected users remains undisclosed.
The case illustrates ongoing difficulties in preventing sophisticated impersonation attacks within mobile application marketplaces.
Why This Matters: Seed phrase theft converts user funds into irreversible losses within minutes, underscoring the need for stronger verification layers in cryptocurrency applications.
UK Lords Set to Vote on Social Media Ban with Digital ID Checks
- The House of Lords scheduled a vote tonight on legislation restricting social media access.
- The bill includes provisions for mandatory digital identity verification for platform users.
- Privacy advocates argue the measures would impose routine identification requirements on all British citizens.
LONDON, Apr 20 (TNGB) – Members of the House of Lords reportedly prepared to vote this evening on a bill that would impose age-based restrictions on social media platforms while embedding compulsory digital identity checks for broader access. If approved, the legislation would return to the House of Commons for additional debate. Parliamentary records and advocacy briefings indicate the provisions aim to protect younger users from harmful content yet contain limited safeguards against data retention or secondary governmental use of identity information. Critics from civil liberties organizations have highlighted the potential for the system to normalize identification requirements for routine online activities such as accessing news or public forums.
The vote occurs amid parallel government consultations on expanding digital identification infrastructure.
Why This Matters: Approval could require millions of adults to submit personal identification documents simply to participate in everyday digital communication.
Seiko USA Website Defaced as Hacker Claims Customer Data Theft
- The Seiko USA website was defaced over the weekend with ransom demands.
- Attackers claimed they had stolen the entire Shopify customer database.
- No independent confirmation of actual data exfiltration has emerged.
USA, Apr 20 (TNGB) – The Seiko USA website reportedly displayed unauthorized messages after attackers gained control of portions of its Shopify-hosted storefront over the weekend. The defacement included claims that the full customer database containing names, email addresses, telephone numbers, and order histories had been downloaded, accompanied by a demand for ransom to prevent public release or sale. Company representatives have not confirmed whether any data was actually exfiltrated or whether the attackers possessed the claimed information. The site was restored within hours, though customer support lines experienced increased volume from concerned users.
Website defacements often precede or accompany attempts to monetize stolen data on underground markets.
Why This Matters: Even unverified claims of customer data exposure can trigger regulatory inquiries and long-term reputational damage for retailers.
Bluesky Blames Intermittent Outages on Sophisticated DDoS Attack
- Bluesky experienced repeated service disruptions beginning the evening of April 15.
- Engineers attributed the issues to an ongoing sophisticated distributed denial-of-service campaign.
- No evidence of unauthorized access to user data was found during the incident.
USA, Apr 20 (TNGB) – The social platform Bluesky reportedly suffered intermittent outages affecting user feeds, notifications, threads, and search functionality starting around 8:40 p.m. Eastern Time on April 15. Company engineers traced the disruptions to a highly coordinated distributed denial-of-service attack that intensified throughout the following day. Technical teams deployed mitigation measures overnight, and by April 17 the platform reported that core services had stabilized, although some residual effects persisted briefly. Officials stated that forensic examination found no signs of data breaches or unauthorized account access. The company’s chief operating officer publicly confirmed the attack’s sophistication and ongoing nature in status updates.
The episode tested the platform’s resilience during a period of rapid user growth.
Why This Matters: Sustained DDoS attacks against social platforms can disrupt public discourse and accelerate user migration to more established competitors.
Congress Debates Renewal of Expiring Surveillance Law Amid Criticism
- Lawmakers negotiated renewal of Section 702 authorities ahead of the April 30 expiration.
- A short-term extension keeping the program operational until April 30 cleared both chambers and received President Trump’s signature.
- Debate centers on oversight, warrant requirements for U.S. person queries, and government data purchases from brokers.
WASHINGTON D.C., Apr 20 (TNGB) – Congressional committees reportedly continued negotiations over reauthorization of Section 702 of the Foreign Intelligence Surveillance Act, which faced expiration at the end of the month. On April 18 both the House and Senate approved a short-term extension that maintains the authority until April 30, and President Trump signed the measure the same day. The Trump administration has supported a clean reauthorization without additional restrictions, while some lawmakers advocate warrant requirements for queries involving Americans and limits on government acquisition of personal data from commercial brokers. The program permits warrantless collection of communications of foreign targets that frequently include incidental U.S. person data, prompting ongoing questions about effectiveness and oversight mechanisms.
The outcome will determine the legal framework for foreign intelligence collection for the next several years.
Why This Matters: Section 702 remains one of the government’s most significant surveillance authorities, directly affecting how intelligence agencies access communications involving Americans.
UK Government Plans Digital ID Expansion Alongside Social Media Ban Debate
- Members of Parliament will debate social media access restrictions and digital identification measures this week.
- Government proposals include expanded facial recognition surveillance deployment.
- Advocacy groups have organized opposition highlighting risks of mission creep and privacy erosion.
LONDON, Apr 20 (TNGB) – British lawmakers reportedly scheduled debate this week on legislation that would restrict social media access for younger users while advancing broader digital identity verification systems. Parallel government plans call for increased use of facial recognition technology in public spaces as part of public safety initiatives. Civil liberties organizations have mobilized campaigns noting the absence of clear limits on data retention periods or secondary uses by authorities. The combined measures build on earlier consultations that attracted millions of public responses opposing mandatory identification requirements for employment and routine services.
The proposals represent a significant expansion of state capabilities to monitor and verify citizen identity in digital and physical spaces.
Why This Matters: Implementation could normalize routine identification checks for employment, travel, and everyday online activity across the United Kingdom.
Claude Desktop Update Raises Browser Integration and Permission Questions
- Anthropic released updates to its Claude Desktop application that modified browser integration and permission settings.
- Some users reported unexpected permission prompts for browser features even when those browsers were not actively in use.
- The changes prompted discussions about data handling transparency and local system access.
USA, Apr 20 (TNGB) – Anthropic’s Claude Desktop application reportedly received updates that altered how the program interacts with browser integrations and system permissions on user devices. Several users documented changes that included permission requests or configuration adjustments affecting browser-related features, even in cases where those browsers were not the primary or actively installed applications on the device. Company documentation described the modifications as enhancements to improve compatibility and security, yet the automatic nature of the changes led to questions about the scope of data access and whether users received adequate advance notice. No evidence of unauthorized data transmission has been reported, though privacy researchers have called for clearer disclosure of how the application handles local system resources.
The incident highlights the challenges of balancing seamless user experience with transparent permission management in AI desktop tools.
Why This Matters: Automatic permission adjustments in productivity applications can erode user confidence and raise legitimate questions about data minimization practices.
British Scattered Spider Hacker Pleads Guilty to Crypto Theft Charges
- British national Tyler Buchanan pleaded guilty in a California federal court to cryptocurrency theft conspiracy charges.
- The charges stemmed from his role in operations linked to the Scattered Spider hacking collective.
- Sentencing is scheduled for later in 2026.
CALIFORNIA, Apr 20 (TNGB) – British national Tyler Buchanan reportedly pleaded guilty on or around April 17-18 in a U.S. federal court in California to charges of conspiracy to commit cryptocurrency theft. Prosecutors linked Buchanan’s activities to the Scattered Spider collective, which has conducted multiple high-value attacks against cryptocurrency exchanges and individual wallets. Court records indicate the plea agreement covers his participation in coordinated campaigns that compromised digital asset platforms and facilitated the theft of significant cryptocurrency holdings. Additional details regarding the full scope of the conspiracy and any cooperation with authorities are expected to surface during the sentencing hearing later this year.
The guilty plea advances U.S. law enforcement efforts against one of the most active cybercrime groups targeting financial services.
Why This Matters: Prosecutions of Scattered Spider members in U.S. courts demonstrate international cooperation in disrupting ransomware and cryptocurrency theft networks that operate across borders.
Chipmakers on Pace to Meet Only 60 Percent of AI Memory Needs by 2027
- Industry forecasts project that memory manufacturers will supply only about 60 percent of anticipated AI demand by 2027.
- Explosive growth in model training and inference workloads continues to exceed production expansion rates.
- New fabrication facilities typically require three to five years to reach meaningful output volumes.
USA, Apr 20 (TNGB) – Semiconductor manufacturers reportedly project they will meet only approximately 60 percent of global demand for high-bandwidth memory required by artificial intelligence systems by 2027. The rapid scaling of large language model training and real-time inference applications has driven consumption rates far beyond current and near-term production capacity. Analysts note that constructing and qualifying advanced fabrication plants for memory production generally takes three to five years from initial investment to volume output. The resulting shortfall is expected to constrain deployment schedules and increase costs for organizations building next-generation AI infrastructure across multiple sectors.
Persistent memory supply constraints represent a fundamental bottleneck for the continued expansion of artificial intelligence capabilities.
Why This Matters: Hardware shortages can delay AI project timelines and raise costs for enterprises, governments, and research institutions investing in the technology.


