AI-Assisted Intrusions, Active Exploitation and Security Industry Moves Feature in Weekly Roundup

· Original article ↗

Summary

The roundup covers stolen AI API keys, an AI-agent PaperCut campaign, fake helpdesk and passkey lures, actively exploited Cisco and GitLab flaws, and security industry and regulatory developments.

Key points

  • Anthropic reported criminals reused stolen customer API keys in intrusions; selected cases also included AI use by suspected ShinyHunters affiliates and a Russian espionage actor.
  • GreyNoise traced an operator using hundreds of AI agents across at least 440 compromised PaperCut instances; domain-admin access was confirmed at 12 organizations.
  • Microsoft described fake helpdesk calls and messages used to lure employees into passkey, MFA, or SSO setup flows and steal cloud access.
  • Cisco confirmed exploitation of an unauthenticated email-gateway flaw that can enable root command execution; on-premises customers need fixed software.
  • GitLab patched an unauthenticated file-read flaw, while researchers also reported exploitation of other GitLab authentication and privilege-escalation flaws.
  • The Cyber Resilience Act’s reporting obligations took effect, requiring covered manufacturers to provide an early warning within 24 hours and a notification within 72 hours for specified security events.
  • Other coverage included AI security debates, security product and funding announcements, and research showing a physical attack against Intel TDX’s default integrity mode.

Article Details

Event Type
Weekly cybersecurity news roundup covering reported AI-assisted intrusions, exploitation of software vulnerabilities, targeted spyware, and security product and company developments.
Impact
Anthropic reported criminals reusing stolen customer API keys in further intrusions, including cases involving suspected ShinyHunters affiliates and a Russian espionage actor. GreyNoise attributed compromises of at least 440 PaperCut instances across 395 organizations in 48 countries to an operator using AI agents, with domain-admin access confirmed at 12 organizations. Microsoft described helpdesk impersonation and passkey or MFA setup lures used to gain cloud access and collect data. Cisco, GitLab, and other products were also reported affected by active exploitation; specific impacts and remediation details vary by incident.

MITRE ATT&CK

CVE

People

Threat Actors

Malware

Vendors

Products

Tools

Related Articles