Creează un cont pentru a te abona la acest blog și a-i urmări articolele pe pagina Lectură.
-
An active malware campaign used typosquatted skills, trojanized packages, and remote loaders to target users of popular AI tools.
-
We spent the last 6 months building a global network of honeypots with exposed AI inference and agent endpoints.
-
AI is incredible.
-
Yesterday (Feb 17, 2026, 12:18AM ET) Cline released an advisory about an unauthorized npm publication.
-
Edit (2/19 2:30AM ET): This blog post was written during an ongoing investigation.
-
Over the past year I’ve been asking people the same question over and over again: when our AI systems are targeted, will you know?
-
I gave a talk at the AI Agent Security Summit by Zenity Labs on October 8th in San Francisco.
-
Enterprise systems provide a terrible user experience.
-
We’re getting asks for more info about the 0click AI exploits we dropped this week at DEFCON / BHUSA.
-
AWS security blog confirms the attacker gained access to a write token and abused it to inject the malicious prompt.