AI & software security
-
Clop-linked web shell targets Windchill data
A purpose-built web shell targeting PTC Windchill can decrypt stored credentials, map engineering repositories, and run additional code while operating inside the application’s own trust boundary.
-
Resilience shouldn’t be a luxury item: why every organisation must prepare for frontier AI
Mark Molyneux, Field CTO North Europe at Commvault, examines how frontier AI could compress vulnerability response windows and why resilience operations should not depend on access to the most advanced models.
-
Application risk climbs as software output accelerates
Sonatype research finds enterprise applications are accumulating more serious vulnerabilities even as remediation speeds up and AI-era software creation accelerates sharply.
-
AI gains collide with workflow governance fears
Perforce’s latest workflow survey finds strong AI productivity gains alongside persistent concerns over job security, content quality, compliance, and visibility into AI-generated changes.
-
GitLab patches critical GraphQL flaw
GitLab has issued an out-of-band patch for a critical GraphQL vulnerability that could let unauthenticated attackers modify or delete public projects and user data on self-managed instances.
-
Microsoft begins final WMIC removal
Microsoft is moving into the final removal phase for the legacy WMIC command-line utility, forcing remaining scripts and administrative workflows onto supported Windows management interfaces.
-
Ray flaw moves into active exploitation
CISA has added a critical Ray code-injection flaw to its exploited-vulnerability catalogue after evidence of real-world attacks against the distributed computing framework.
-
Windows flaw joins ransomware attack chains
CISA says ransomware operators are exploiting a patched Windows Task Host privilege-escalation flaw, adding another known vulnerability to attack chains against Windows environments.
-
France turns tax breach into AI test
France plans to use sovereign AI providers to probe government cyber weaknesses after attackers stole information concerning roughly 700,000 taxpayers.
-
Fortinet buys Virtue AI for agent security
Fortinet has acquired Virtue AI, adding automated AI testing, runtime controls, agent visibility, and Model Context Protocol scanning to its security platform.