Warnings over AI agent vulnerabilities: Google tracks rise in malicious content as Microsoft urges caution

Google security researchers recorded a relative rise of 32% in malicious vulnerability content online between November 2025 and February 2026, but the company said most of it consists of educational materials, research examples or pranks. In other coverage, Microsoft warned companies against trusting AI agents.

AI-written summary of 2 sources; every verified point is backed by a verbatim quote. How Iyan works

• Google researchers recorded a relative rise of 32% in malicious vulnerability content between November 2025 and February 2026.

  «رصد باحثون أمنيون في غوغل ارتفاعًا نسبيًا بنسبة 32% في محتوى الثغرات الأمنية الخبيث على الإنترنت بين نوفمبر 2025 وفبراير 2026»

  الغد
• Google said most of what it found was educational materials, research examples or pranks, not actual attacks.

  «لكن الشركة قالت إن معظم ما وجدته كان مواد تعليمية أو أمثلة بحثية أو مقالب»

  الغد
• Microsoft issued a warning to companies not to trust AI agents.

  «"مايكروسوفت" تحذر الشركات: لا تثقوا في وكلاء الذكاء الاصطناعي»

  العربية

Iyan — See it for yourself. What changed in your world today, why it matters, and how we know.
How it works • Privacy • Terms