Security news weekly round-up - 28th August 2026

Staying safe online remains a top priority, with users advised to be cautious when clicking links, downloading apps, and sharing personal information. AI can be used to piece together seemingly private data, posing a threat. Attackers impersonate AI brands like Claude to spread malware, often disguising malicious payloads as legitimate files like "claude" or "claude.msixbundle."

Frontier AI labs have not disclosed their plans for containing rogue models. Frontier AI labs have not disclosed their plans for containing rogue models. Meanwhile, the Android malware Frontier AI Labs "ToxicPanda" utilizes VPN permissions to block Google Play and includes a PIN-harvesting module targeting 140 financial and cryptocurrency apps.

In other malware news, AliExpress users discovered they were inadvertently tracked through audio glitches, with researchers identifying a lack of consent for intrusive tracking methods. Interpol reported uncovering a disturbing trend of West African crime rings using sextortion against minors, typically involving social media manipulation and coercing victims into sharing explicit content under threat of public exposure.

Experts caution that even after passing Multi-Factor Authentication (MFA), users should still verify identities to prevent attackers from exploiting legitimate MFA credentials. Furthermore, AI-assisted reconnaissance highlights the risk of personal information inadvertently shared online, potentially impacting both professional and personal lives as the boundary between work and home increasingly blurs in a hybrid work environment.