All news in category “Incidents and Data Breaches”

🔒 Washington Hotel, a business brand of Fujita Kanko Inc., disclosed a ransomware infection after an intrusion on Friday, February 13, 2026 at 22:00 local time. The company says it immediately disconnected affected servers, formed an internal task force, and engaged external cybersecurity experts to assess impact and coordinate recovery; preliminary findings indicate attackers accessed various business data. Customer records are unlikely to have been exposed because those are held by a separate vendor, but some properties experienced operational effects such as temporarily unavailable credit-card terminals.

🔒 Eurail B.V. confirmed that customer data stolen in a breach earlier this year is now being offered for sale on the dark web, and a sample dataset was published on Telegram. The company says it is still determining which specific records and how many customers are affected, but reported compromised fields may include full names, passport and ID numbers, IBANs, health details, and contact information. GDPR-required notifications have been filed and non-EU authorities will be informed. Customers are urged to change reused passwords, monitor bank accounts closely, and contact privacyhelp@eurail.com for support and FAQs.

🔒 Dutch police arrested a 40-year-old man in Ridderkerk after he downloaded confidential documents that an officer mistakenly shared via a download link and then refused to delete them unless he received "something in return." Authorities detained him on suspicion of computer trespass, searched his home and seized storage devices to recover the files. Police reported the breach and are investigating, saying there is no indication the documents were distributed further.

🔓 Hudson Rock says an info‑stealer, likely a Vidar variant, exfiltrated an OpenClaw agent's configuration, including openclaw.json, device.json and soul.md. The files contain gateway tokens, cryptographic keys and the agent's operational 'soul,' which could let attackers impersonate the AI assistant or connect to local instances if exposed. The incident signals a shift from stealing credentials to harvesting AI agent identities, and vendors should expect targeted modules to follow.

🔐 Hudson Rock has observed information-stealing malware exfiltrating configuration and memory files from the OpenClaw agent framework, exposing API tokens, private keys, and persistent agent memory. The activity, attributed to a Vidar-like infostealer and recorded on 13 February 2026, captured openclaw.json, device.json, and agent 'soul' and memory files. With these items an attacker could impersonate the device, bypass Safe Device checks, access encrypted logs, or fully compromise a user's digital identity. Organizations should audit agent directories, apply vendor fixes, and enforce strict filesystem permissions immediately.

🔒 SOCRadar has uncovered a phishing campaign named Operation DoppelBrand that targeted Fortune 500 financial, insurance and technology firms between December 2025 and January 2026. The activity is attributed to financially motivated actor GS7 and relies on lookalike domains and cloned login portals to harvest credentials, which are forwarded to Telegram bots. Successful compromises often result in the deployment of legitimate remote access tools such as LogMeIn Resolve, delivered via MSI installers and supported by VBS loaders for privilege escalation and silent installation.

🔍 An estimated 37 million global installs of Chrome extensions have been found transmitting users’ browsing histories to external servers. Independent researcher 'Q Continuum' identified 287 extensions that sent data closely matching visited URLs during automated simulated browsing. Flagged add-ons spanned VPNs, productivity tools, shopping/coupon helpers and browser utilities, and many obfuscated outbound payloads using base64, ROT47, compression or strong encryption. The researcher warned such exfiltration could expose internal corporate URLs and, where cookies or session data are accessible, enable credential harvesting.

🔒 Odido, the largest mobile operator in the Netherlands, disclosed a data breach affecting its customer contact system and potentially impacting up to 6.2 million people. While the company says no passwords, call records or billing data were taken, exposed fields reportedly include names, home and email addresses, IBANs, dates of birth and passport/driver's license numbers. Odido has contained the intrusion, engaged external cybersecurity experts and will contact affected customers directly.

🔒 Ten years after the February 2016 operation that attempted to steal $951 million via fraudulent SWIFT messages, the Bangladesh Bank heist remains a defining case for cyber resiliency. Attackers attributed to the Lazarus Group used spear-phishing, backdoors, keyloggers and printer sabotage to capture credentials and erase audit trails, enabling 35 fraudulent transfer attempts. The incident exposed basic control failures—lack of network segregation, exposed SWIFT systems, and limited endpoint monitoring—and helped drive mandatory measures such as the SWIFT Customer Security Program.

🔍 Canada Goose is investigating after data extortion group ShinyHunters published an archive claiming more than 600,000 customer records tied to past transactions. The 1.67 GB JSON dataset reportedly contains names, emails, phone numbers, billing and shipping addresses, IPs, order histories, and partial payment card data (brands, BINs, last four digits). Canada Goose says it has found no evidence of a breach of its own systems and that no unmasked financial data appears present, while it reviews the dataset to verify accuracy and scope.

⚠️ Microsoft has identified a novel ClickFix social-engineering variant that instructs victims to run an nslookup against an attacker-controlled resolver to retrieve a malicious PowerShell script embedded in the DNS NAME field. The response is parsed and executed via cmd.exe, then pulls a second-stage ZIP containing a Python runtime and scripts that lead to the ModeloRAT remote-access trojan. Organizations should monitor unusual DNS queries to untrusted nameservers and apply endpoint controls to block unauthorized script execution and persistence.

🚨 Threat actors are abusing Pastebin comments to promote a ClickFix-style social engineering campaign that tricks cryptocurrency users into executing JavaScript in their browser, enabling attackers to hijack Bitcoin swap transactions on Swapzone.io. Victims are directed to copy a javascript: snippet from a hosted paste and execute it in the address bar; the injected, obfuscated payload overrides the exchange's swap logic and replaces deposit addresses with attacker-controlled wallets. The code also tampers with displayed rates and offers to simulate successful arbitrage. Because the script runs within the victim's authenticated session, the interface looks legitimate while funds are irreversibly redirected to attackers.

🛡️ GreyNoise telemetry indicates a single IP hosted by PROSPERO OOO is responsible for roughly 83% of active exploitation attempts against Ivanti Endpoint Manager Mobile (EPMM), targeting CVE-2026-21962 and CVE-2026-24061. Between Feb 1–9 researchers observed 417 exploit sessions from eight source IPs, with a sharp spike on Feb 8. Activity appears automated, using OAST-style DNS callbacks consistent with initial access broker behavior; Ivanti has released hotfixes and will issue full patches in Q1.

📬 Cybercriminals are mailing phishing letters impersonating Trezor and Ledger to trick hardware wallet owners into surrendering recovery phrases. The letters pressure recipients with deadlines for an Authentication Check or Transaction Check and instruct them to scan QR codes that lead to cloned setup pages. Those pages prompt entry of 24-, 20- or 12-word seed phrases, which are then sent to attacker-controlled servers, allowing funds to be stolen. Never share your recovery phrase; manufacturers will never ask for it.

⚠️ A new variant of a fake recruiter campaign attributed to North Korean actors is targeting JavaScript and Python developers with cryptocurrency-themed coding tasks. Attackers publish seemingly legitimate job projects and embed malicious dependencies on npm and PyPI that install a remote access trojan reported as Graphalgo. The operation is modular and resilient, with 192 malicious packages identified and tactics such as delayed activation and token‑protected command channels. Affected developers are advised to rotate tokens and passwords and to reinstall compromised systems.

⚠️ Threat actors are abusing public Claude artifacts and manipulated Google Search results to trick macOS users into running malicious Terminal commands. These commands download and execute a loader that installs the MacSync infostealer, which harvests keychain data, browser credentials, and crypto wallets, then exfiltrates the data to a hardcoded command-and-control server. Researchers warn users not to run unverified shell commands and to verify safety before executing them.

🔒 South Korea's Personal Information Protection Commission fined Louis Vuitton, Christian Dior Couture, and Tiffany a combined $25 million after cloud-based customer management systems were compromised, exposing data for more than 5.5 million customers. Investigators found an employee device infected with malware at Louis Vuitton and successful phishing and voice-phishing attacks at Dior and Tiffany that granted attackers access to the SaaS platform. Regulators cited failures to enforce IP-based access controls, deploy strong authentication, restrict bulk downloads, and monitor access logs, and penalized late breach notification. The PIPC emphasized that using a SaaS provider does not relieve companies of responsibility for protecting client data.

⚠️ Google Threat Intelligence Group (GTIG) attributes a previously undocumented actor, likely linked to Russian intelligence, to campaigns using CANFAIL against Ukrainian defense, military, government, and energy organizations. The actor has expanded interest to aerospace, defense-adjacent manufacturing, nuclear and chemical research, and humanitarian groups, often impersonating Ukrainian and Romanian energy firms in phishing. Operators used LLMs to produce reconnaissance and social-engineering lures, embedding Google Drive links to RAR archives that deliver obfuscated JavaScript which spawns PowerShell memory-only droppers. GTIG links this activity to the PhantomCaptcha campaign disclosed by SentinelOne SentinelLABS in October 2025.

🔎 Google Threat Intelligence Group (GTIG) warns that state-sponsored actors from China, Iran, Russia, and North Korea are conducting sustained cyber operations against the defense industrial base (DIB). GTIG highlights four themes: targeting battlefield technologies like drones, exploiting hiring and personnel processes, leveraging edge devices for initial access, and capitalizing on manufacturing supply chain breaches. Observed tactics include bespoke malware families, abuse of secure messaging linking, careful endpoint-evasion techniques, and use of relay networks to complicate detection and attribution.

🔍 Cisco Talos reports that threat actor UAT-9921 has deployed the modular VoidLink framework in campaigns targeting technology and financial organizations. The post-compromise toolkit—built in Zig, C, and Go—supports compile-on-demand plugins, stealthy persistence, and runtime evasion. Operators install SOCKS proxies and use open-source scanners for internal reconnaissance and lateral movement, and evidence suggests a Windows implant and role-based access controls are present.