Cybersecurity Brief – 2025-10-11

Sat, 11 Oct 2025 00:00:00 +0000

Major Incidents or Breaches

The FBI has seized all domains associated with BreachForums, operated by the ShinyHunters group, which was used as a portal for leaking corporate data obtained through ransomware and extortion attacks, including recent Salesforce-related incidents. Despite the takedown, ShinyHunters continues to issue extortion threats against Salesforce victims.
Two AI companion applications exposed millions of private chat records from over 400,000 users due to inadequate security controls.
A critical zero-day vulnerability (CVE-2025-11371) in Gladinet CentreStack and TrioFox file sharing products is being actively exploited in the wild, allowing unauthenticated local attackers to access system files.
Sophisticated malware has been deployed in Oracle E-Business Suite (EBS) zero-day attacks, with exploitation believed to have started as early as July 10 and affecting dozens of organizations.
A large-scale DDoS attack was launched by the Aisuru botnet, leveraging compromised IoT devices hosted on major US ISPs.
Microsoft has observed a threat actor, Storm-2657, hijacking HR SaaS accounts to divert employee salary payments to attacker-controlled accounts; US universities have been specifically targeted in these payroll diversion attacks.

Fortra disclosed details of active exploitation of CVE-2025-10035, a critical vulnerability in GoAnywhere Managed File Transfer (MFT).
13 unpatched vulnerabilities in Ivanti Endpoint Manager have been publicly released by the Zero Day Initiative (ZDI), permitting remote code execution and privilege escalation.
Juniper Networks released patches for over 200 vulnerabilities in Junos Space and Junos Space Security Director, including nine critical-severity flaws.
175 malicious npm packages, collectively downloaded over 26,000 times, have been used in a credential phishing campaign.
Security researchers have detailed an active Stealit malware campaign abusing the Node.js Single Executable Application (SEA) feature, distributed via game and VPN installers.

North Korean threat actors have expanded their fraudulent remote employment schemes to target nearly every industry hiring remote workers, including architectural design roles, using fake identities and documentation.
Chinese APT group Storm-2603 is abusing the Velociraptor digital forensics and incident response (DFIR) tool to gain persistent access to victim networks and facilitate ransomware deployment.
A coordinated campaign has been identified targeting Cisco, Fortinet, and Palo Alto Networks devices, with attacks originating from shared infrastructure.
The RondoDox botnet employs a “shotgun” approach, using more than 50 exploits against unpatched routers, DVRs, NVRs, CCTV systems, servers, and other network devices globally.

AI-generated deepfakes are proliferating, with OpenAI releasing an updated AI video and audio generation system, raising concerns about the acceleration and accessibility of deepfake creation.
The security industry is experiencing inadequate investment in deepfake detection, despite high organizational awareness and frequent encounters with AI-augmented threats.
Security Operations Centres (SOCs) are increasingly integrating AI-powered agents, with projections indicating significant transformation in SOC workflows by 2026.
Microsoft has added agentic AI capabilities to its Sentinel platform, previewing new security graph and MCP server features.
Attackers are leveraging malicious npm packages and abusing legitimate software features (e.g., Node.js SEA) to distribute malware and harvest credentials.
1Password addressed a critical security gap in its AI browser agent, which posed risks of credential leakage.
Timely patch management is highlighted as a key defense against escalating ransomware attacks on manufacturing and industrial control system (ICS) environments.

Apple has expanded its bug bounty program, doubling the maximum payout to $2 million for zero-click remote code execution vulnerabilities, with potential bonuses raising the maximum to $5 million. The program now includes new research categories and a more transparent reward structure.
Microsoft reminded users that Windows 11 23H2 Home and Pro editions will reach end of support in 30 days, after which security updates will cease.
Google Chrome will automatically revoke notification permissions for inactive websites to reduce user alert fatigue.
Apple voiced concerns over proposed age-verification laws, warning that increased data collection for compliance could heighten user privacy risks.
NIST is revising password guidance, indicating that excessively complex password requirements may no longer be necessary.