中文 English Français Русский Español العربية

Anthropic Says Russian Hackers Used Claude AI to Automate Malware Evasion

Eduard Kovacs SecurityWeek 2026-09-11 16:47 20 views

Anthropic reveals how criminal groups are increasingly targeting AI vendors' own infrastructure, including to steal a pre-release Claude model. The post Anthropic Says Russian Hackers Used Claude AI to Automate Malware Evasion appeared first on SecurityWeek.

Read original: https://www.securityweek.com/anthropic-says-russian-hackers-used-claude-ai-to-automate-malware-evasion/

My Favorites →

← Previous GitHub’s August Outages Show Growth Is Outpacing Infrastructure Next → A practical startup budget for React Native apps

Designing a Privacy-First Architecture for Sensitive Data: Zero-IP Logging, Postgres RLS, and AI Safety

When building software that handles sensitive human data (like workplace feedback, satisfaction surveys, and compliance reports), security and privacy cannot be treated as optional features or mere legal disclaimers. They must be embedded into the core system design. At Sigilo Profissional (sigilop

Security 09-15 65 Dev.to (EN Zone)

Patch Rollback Risk: When Recovery Reintroduces the Exposure You Just Closed

Patch rollback risk is the operational reality sitting underneath Microsoft's September 2026 Remote Desktop Services failure: the fastest way to restore a broken service can be the exact action that reopens a vulnerability the patch was written to close. The September 8 cumulative updates — KB512

Security 09-15 47 DEV Community

Emacs arbitrary code execution flaw

Sean Whitton has announced that the original fix for an arbitrary code execution flaw in Emacs (CVE-2024-53920) was incomplete. Bas Alberts discovered that viewing or editing untrusted files in modes other than Emacs's Lisp mode can also result in arbitrary code execution. This problem affects all

Security 09-14 57 LWN.net

I built a chat app that forgets 🔥

There's a kind of conversation that doesn't belong in WhatsApp. 💬 You need to send someone a document. Or say something that shouldn't sit in a thread forever. Or talk to a person you'd rather not add to your contacts. Every option feels wrong — message them somewhere that keeps a copy of everythin

Security 09-14 47 Dev.to (EN Zone)

Comments0

No comments yet — be the first