Transparency Under Scrutiny
A security researcher discovered code inside Anthropic’s AI coding assistant Claude Code that secretly collected information about users’ time zones and proxy usage.
Links to Known Threat Actors
Microsoft has issued a warning about GigaWiper, a Windows backdoor that combines advanced espionage capabilities with destructive data wiping while disguising its activity as a ransomware attack.
Complete Data Destruction
Security researchers at Sysdig have documented what they describe as the first fully autonomous ransomware attack executed by an AI agent.
Sandbox Escape
Security researchers have uncovered two critical vulnerabilities in the AI development environment Cursor. By using specially crafted prompts, attackers can escape the application’s sandbox and execute code directly on the host operating system — without requiring any user interaction.
Social Engineering
A new macOS Malware campaign uses ClickFix social engineering and Terminal commands to silently download malicious files, steal passwords and compromise cryptocurrency data.
Supply Chain Attacks increase
A new ranking reveals that hacker groups are increasingly targeting supply chains and relying on rented infrastructure with market shares of up to 89 percent.
AI agent–driven attack campaign
A hacker compromised 14 companies using Anthropic’s Claude Code agent and OpenAI’s Codex. The attacker’s lack of technical expertise was compensated for by the AI.
Vulnerability in Oracle Software
Estée Lauder reports a data breach. Attackers exploited a vulnerability in Oracle E-Business Suite to steal personal and customer data.
Infostealer malware
An unprotected server has exposed 24 billion login credentials in plaintext, raising major concerns over the security of billions of online accounts. According to Cybernews, billions of accounts without multi-factor authentication (MFA) are at risk.
Most Read Articles
26. August 2026
20. August 2026
17. August 2026
12. August 2026
10. August 2026