Lantronix EDS5000 devices are vulnerable to remote code execution via command injection in the login protocol (CVE-2025-67038, CVSS 9.8), and active exploitation is occurring.
A large-scale alliance of established technology companies and financial institutions pools resources to coordinate remediation of open-source security gaps in response to AI-powered vulnerability discovery.
PostgreSQL contains multiple critical vulnerabilities allowing remote code execution and data manipulation; BSI classifies the threat level as elevated.
Cybercriminals are split on AI adoption: while some embrace automation and efficiency gains, others fear their existing illegal services are endangered by AI-powered security.
The parallel activity of two independent ransomware groups on the same SharePoint servers demonstrates that attackers are increasingly conducting overlapping campaigns, requiring centralized visibility across all layers.
Summer holiday absences lead to 69 percent email fraud between June and August, but ransomware remains undetected an average of nine days longer, with damages becoming visible only in autumn.
Following a rail radio outage, security politicians are calling for a statutory ban on Chinese components in critical infrastructure to prevent sabotage and espionage.
Autonomous AI agents require observability platforms that make decision-making fully traceable, display costs transparently, and enforce defined action boundaries.
Employees unknowingly enter sensitive data into unauthorized AI services; traditional DLP solutions fail to capture these new data paths and require context-based risk analysis instead of blanket blocks.
