Skip to main content

TrendAI™
Blog de Segurança

Blog de Segurança TrendAI

Blogs em Destaque

LLMs
When Tokenizers Drift: Hidden Costs and Security Risks in LLM Deployments

A tokenizer lies at the core of every large language model. When it drifts, whether from unseen flaws or adversarial interference, costs rise and performance drops. We explore this emerging risk, its implications, and the measures to prevent it.

Read Article
Cyber crime Malware
Fast, Broad, and Elusive: How Vidar Stealer 2.0 Upgrades Infostealer Capabilities

TrendAI™ Research examines the latest version of the Vidar stealer, which features a full rewrite in C, a multithreaded architecture, and several enhancements that warrant attention. Its timely evolution suggests that Vidar is positioning itself to occupy the space left after Lumma Stealer’s decline.

Read Article
Shifts in the Underground: The Impact of Water Kurita’s (Lumma Stealer) Doxxing

A targeted underground doxxing campaign exposed alleged core members of Lumma Stealer (Water Kurita), resulting in a sharp decline in its activity and a migration of customers to rival infostealer platforms.

Read Article
Operation Zero Disco: Attackers Exploit Cisco SNMP Vulnerability to Deploy Rootkits

Trend™ Research has uncovered an attack campaign exploiting the Cisco SNMP vulnerability CVE-2025-20352, allowing remote code execution and rootkit deployment on unprotected devices, with impacts observed on Cisco 9400, 9300, and legacy 3750G series.

Read Article
Cyber threats Exploits & Zero-Days IoT & smart devices
RondoDox: From Targeting Pwn2Own Vulnerabilities to Shotgunning Exploits

The Trend Zero Day Initiative™ (ZDI) and Trend™ Research teams have identified a large-scale RondoDox botnet campaign exploiting over 50 vulnerabilities across more than 30 vendors, including flaws first seen in Pwn2Own contests.

Read Article
Malware Phishing & BEC
Malware Autopropagável se Espalha Via WhatsApp, Alvo: Usuários Brasileiros

A TrendAI™ Research identificou uma campanha ativa se espalhando via WhatsApp através de um anexo de arquivo ZIP. Quando executado, o malware estabelece persistência e sequestra a conta do WhatsApp comprometida para enviar cópias de si mesmo para os contatos da vítima.

Read Article
Ransomware & extortion
Novo LockBit 5.0 Alvo Windows, Linux, ESXi

A TrendAI™ Research analisou binários de origem da atividade mais recente do notório Ransomware LockBit com sua versão 5.0, que exibe técnicas avançadas de ofuscação, anti-análise e capacidades multiplataforma perfeitas para sistemas Windows, Linux e ESXi.

Read Article
Cyber crime
What We Know About the NPM Supply Chain Attack

TrendAI™ Research outlines the critical details behind the ongoing NPM supply chain attack and offers essential steps to stay protected against potential compromise.

Read Article
AI
Operadores do EvilAI Usam Código Gerado por IA e Aplicativos Falsos para Ataques de Grande Alcance

Combinando código gerado por IA e engenharia social, os operadores do EvilAI estão executando uma campanha em rápida expansão, disfarçando seu malware como aplicativos legítimos para contornar a segurança, roubar credenciais e comprometer persistentemente organizações em todo o mundo.

Read Article
Ransomware & extortion
Unmasking The Gentlemen Ransomware: Tactics, Techniques, and Procedures Revealed

An analysis of the Gentlemen ransomware group, which employs advanced, adaptive tactics, techniques, and procedure to target critical industries worldwide.

Read Article