InfoNewsLLM-specific
Top GenAI security resources — October 2026
- Published
- Record updated
Summary
Anthropic's September 2026 testing found that the open-weight GLM-5.3 loses its safeguards under prefilled reasoning and abliteration, reaching full engagement with malicious cyber requests and writing working exploits. A jailbreak planted in Kimi's custom memories survived session restarts, and an SGLang pickle allowlist flaw let an unauthenticated request reach os.system, unpatched at disclosure.
Related items
- InfoGemini 4 Argon for RoboticsSame vendor · DeepMind Safety Research
- InfoGemini 4 Argon: Built to solve your hardest problemsSame vendor · DeepMind Safety Research
- InfoGoogle finally releases Gemini 4 ArgonSame vendor · The Verge (AI)
- InfoCompanies rushed in on AI. Now some are pulling backSame vendor · The Guardian Technology
- InfoAnthropic to disclose new revenue numbers as AI lab financials become central to tech tradeSame vendor · CNBC Technology