Hackers Leveraging Multiple AI Services to Compromise 600+ FortiGate Devices
Between Wed, Jan 11, 2026, and Thu, Feb 18, 2026, a financially-driven threat actor compromised over 600 FortiGate devices in more than 55 countries by exploiting commercial generative AI services.
Between Wed, Jan 11, 2026, and Thu, Feb 18, 2026, a financially-driven threat actor compromised over 600 FortiGate devices in more than 55 countries by exploiting commercial generative AI services.
The attack demonstrated that AI could significantly lower the technical barriers for conducting cyber operations. The threat actor utilized credential-based exploitation of FortiGate management interfaces exposed online, without deploying zero-day vulnerabilities or novel techniques.
Ports 443, 8443, 10443, and 4443 were systematically scanned to identify appliances with weak or reused credentials using single-factor authentication. Extracted FortiGate configuration files contained high-value data, such as SSL-VPN user credentials, administrative credentials, network topology data, IPsec VPN configurations, and firewall policies. These were processed using AI-assisted Python scripts to facilitate large-scale credential harvesting.
The targeting was opportunistic and automated mass scanning was used. Amazon Threat Intelligence noted patterns of compromise at the organizational level, particularly within managed service provider environments. Compromised devices were notably concentrated in South Asia, Latin America, the Caribbean, West Africa, Northern Europe, and Southeast Asia.
The threat actor used at least two commercial large language models (LLMs) during operations. One LLM was employed for tool development and attack planning, while the other assisted in network pivoting within compromised networks. A documented case revealed that the actor submitted a victim's network topology to an AI service for lateral movement guidance.
The post-exploitation phase involved deploying Meterpreter with the Mimikatz module to execute DCSync attacks against domain controllers, extracting NTLM credential databases from multiple Active Directory environments . Lateral movement was achieved through pass-the-hash, pass-the-ticket, and NTLM relay attacks. Veeam Backup & Replication servers were specifically targeted to disrupt recovery capabilities before ransomware deployment.
Despite the operation's scale, Amazon's analysis identified consistent skill limitations, with the actor abandoning targets with effective defenses. The AI-generated reconnaissance framework, written in Go and Python, exhibited unsophisticated development traits.
The attack demonstrated that AI could significantly lower the technical barriers for conducting cyber operations.
CVE ID Affected Product CVSS Score Description
CVE-2019-7192 FortiOS 9.8 Path traversal allowing unauthenticated credential access
CVE-2023-27532 Veeam Backup & Replication 7.5 Unauthenticated API access for credential extraction
CVE-2024-40711 Veeam Backup & Replication 9.8 Remote Code Execution via deserialization flaw
Amazon shared indicators of compromise with industry partners to coordinate disruption efforts. Organizations using FortiGate appliances are advised to:
Remove management interfaces from internet exposure. Enforce multi-factor authentication for all VPN and administrative access. Rotate SSL-VPN and administrative credentials. Audit Active Directory for DCSync activity (Event ID 4662).
Given the campaign's reliance on legitimate open-source tools, it is recommended to monitor for anomalous VPN authentication patterns, unexpected Active Directory replication, and unauthorized PowerShell module loading on backup servers.
IOC Value IOC Type First Seen Last Seen Annotation
212[.]11[.]64[.]250 IPv4 Jan 11, 2026 Feb 18, 2026 Threat actor infrastructure used for scanning and exploitation operations
185[.]196[.]11[.]225 IPv4 Jan 11, 2026 Feb 18, 2026 Threat actor infrastructure used for threat operations
Based on reporting by Cyber Security News.