GTIG AI Threat Tracker: Distillation, Experimentation, and (Continued) Integration of AI for Adversarial Use | Google Cloud Blog

GTIG observed widespread misuse of generative AI in late 2025, including an uptick in model extraction (“distillation”) attempts and AI-augmented operations such as reconnaissance, hyper-personalized phishing, and AI-assisted malware development. Notable examples include the HONESTCUE downloader that called Gemini’s API to generate stage-two code and the COINBAIT phishing kit built with AI-assisted code generation and hosted on legitimate services (#HONESTCUE #COINBAIT)

Read More
Turning IBM QRadar Alerts into Action with Criminal IP

Criminal IP is now integrated with IBM QRadar SIEM and QRadar SOAR, bringing AI-powered, IP-based threat intelligence directly into QRadar detection, investigation, and response workflows. The integration provides real-time IP risk scoring from firewall logs, in-context investigations inside QRadar, and automated SOAR enrichment to speed prioritization and response. #CriminalIP #IBMQRadar

Read More
Claude LLM artifacts abused to push Mac infostealers in ClickFix attack

Threat actors are abusing public Claude artifacts and malicious Google Ads in ClickFix campaigns to trick macOS users into pasting shell commands that install the MacSync infostealer. Researchers from Moonlock Lab and AdGuard observed multiple variants and thousands of views, with the same C2 infrastructure linking the activity to a single actor. #MacSync #Claude

Read More
Fake job recruiters hide malware in developer coding challenges

A new variation of a fake recruiter campaign attributed to North Korea’s Lazarus group targets JavaScript and Python developers with cryptocurrency-related coding tasks that trick applicants into running malicious repositories. Researchers uncovered 192 malicious npm and PyPI packages dubbed “Graphalgo” that deliver a modular RAT capable of MetaMask theft, token‑protected C2, remote command execution, and data exfiltration; impacted developers should rotate credentials and reinstall their OS. #Graphalgo #Lazarus

Read More
Malicious Chrome Extension Steals Meta Business Manager Exports and TOTP 2FA Seeds

Socket Threat Research discovered a malicious Chrome extension named CL Suite by @CLMasters that advertises Meta Business Suite scraping and 2FA generation while exfiltrating TOTP seeds, current 2FA codes, Business Manager contact CSVs, and analytics to threat actor infrastructure. The extension reports data to getauth[.]pro (and optionally forwards payloads to a Telegram channel), undermining 2FA and enabling account takeover and long-term business asset hijacking. #CLSuite #MetaBusinessSuite

Read More
Dark Web Profile: The Gentlemen Ransomware

The Gentlemen is an operationally disciplined ransomware group first observed in mid-to-late 2025 that conducts double‑extortion attacks across Windows, Linux, NAS, BSD, and ESXi environments using password‑protected, operator-driven builds. Their campaigns leverage exposed internet-facing services and compromised administrative credentials, and victims have been publicly listed on a Dark Web leak site. #TheGentlemen #ESXi

Read More
Governor Confirms Virgin Islands EDA Cyberattack; Official Statement Pending

The U.S. Virgin Islands Economic Development Authority (EDA) has been reported as the target of a cyberattack, and agency officials have not yet commented publicly. The governor confirmed the incident, said an official statement is forthcoming, and authorities warned that cyber ransom attacks are increasing locally while offering assistance for any security breaches. #EDA #USVIEDA

Read More
OysterLoader Unmasked: The Multi-Stage Evasion Loader

OysterLoader (aka Broomstick / CleanUp) is a multi-stage C++ loader distributed via fake signed MSI installers that delivers payloads (notably Rhysida ransomware and commodity stealer Vidar) using staged shellcode, custom LZMA, and steganographically hidden DLLs. Its operators use extensive obfuscation (API-hammering, dynamic API hashing, custom Base64 alphabets and RC4), robust HTTP-based C2 with fallback servers, scheduled-task persistence, and anti-analysis checks. #OysterLoader #Rhysida

Read More
Fake recruiter campaign targets crypto devs

ReversingLabs uncovered a modular software‑supply‑chain campaign called graphalgo, attributed to North Korea’s Lazarus Group, that targets JavaScript and Python developers via fake recruiter job tasks and malicious packages on npm and PyPI. The operation uses fake company personas (e.g., Veltrix Capital), social outreach (LinkedIn, Facebook, Reddit), dependency-based infection (packages such as bigmathutils and graphnetworkx), and multistage downloaders that deploy a token‑protected RAT communicating with codepool[.]cloud. #LazarusGroup #graphalgo

Read More
LummaStealer Is Getting a Second Life Alongside CastleLoader

Bitdefender analysis shows LummaStealer has rapidly rebounded after a 2025 takedown by rebuilding infrastructure, leveraging social-engineering lures and multiple loaders — most prominently CastleLoader — to deliver in-memory payloads and exfiltrate credentials, cookies, documents and crypto data. Detection opportunities include an anomalous DNS lookup pattern from CastleLoader and behavioral indicators around…

Read More
How ClickFix Opens the Door to Stealthy StealC Information Stealer

This analysis details a multi-stage Windows attack that begins with a fake Cloudflare CAPTCHA (ClickFix) social-engineering prompt to trick victims into executing malicious PowerShell, which in-memory loads shellcode, a PE downloader, and ultimately the StealC information stealer. The StealC payload harvests browser credentials, crypto wallets, Steam and Outlook data, system fingerprints, and screenshots exfiltrating data to RC4-encrypted HTTP C2 servers. #StealC #ClickFix

Read More
Leaked technical documents show China rehearsing cyberattacks on neighbors’ critical infrastructure

China appears to have used a secret cyber range called Expedition Cloud to rehearse attacks on replicas of neighboring countries’ critical infrastructure, according to a cache of leaked development and training files. The materials, linked to developer files from CyberPeace and obtained via an exposed FTP server, show staged reconnaissance and…

Read More
Microsoft: Exchange Online flags legitimate emails as phishing

Microsoft is investigating an ongoing Exchange Online issue that began on February 5, where a new URL rule is incorrectly flagging legitimate emails as phishing and quarantining them, blocking some users from sending or receiving messages. The company is reviewing and releasing quarantined messages while working to unblock affected URLs and confirm full remediation. #Microsoft #ExchangeOnline

Read More