Skip to content
iThome (Taiwan)· 羅正漢·Indexed by this site · Originally published· Event →· Its place in the scan record →SelectedAttention41

A security vendor reveals the ARTEX attack incident response process: the AI agent launched the attack within minutes of obtaining an API key, and the defender used AI to accelerate the investigation

資安業者揭露ARTEX攻擊事件應變過程,AI代理程式取得API金鑰後幾分鐘內即發動攻擊,防守方也用AI加速調查

AI summary

Security vendor Eye Security disclosed an incident response case in which an enterprise was attacked by the AI agent attack framework ARTEX; the attacker launched the attack within minutes of obtaining an API key, and the defender also used AI to accelerate the investigation. The attacker exploited a vulnerability in the enterprise's PHP file upload function to plant a web shell, stole keys from environment variables, and upon hitting API rate limiting rewrote the program within about 3 minutes to bypass it, raising traffic to over 9,000 requests per minute; after being blocked, the attacker switched to over 17,000 residential proxy IPs; the attacker also used Git repositories and automated deployment mechanisms to deploy malicious Pods in Kubernetes, which were rebuilt within one second of deletion. Eye Security assesses that the attack aimed to obtain cashable data such as gift card codes rather than ransom. The defender restricted the AI agent to read-only access and collaboratively analyzed logs via a private Git repository, checking EDR telemetry every 30 minutes; several days later the enterprise restored service and no further intrusion was observed.

Why it was selected

This case fully presents the incident response process for an AI agent attack; the defender's approach of using read-only-permission agents to analyze logs in parallel can serve as a reference for incident response teams.

Read originalithome.com.tw