The challenge of AI in cloud security
summary
Attackers exploit a known file‑sharing vulnerability to get a bash reverse shell using reverse TCP, download a uniquely tagged ransomware binary called iFire plus a demo script, and encrypt or exfiltrate cloud data in seconds. Detection uses multi-engine AI — including behavioral AI that matches atomic MITRE ATT&CK behaviors and a ML classifier for file attributes — to sever SSH, quarantine files, and stop the campaign. The org also shows a generative assistant (Purple AI) and a STAR correlation engine tying endpoint, cloud and identity data for faster triage.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:15Agenda
- 2:18Data points
- 5:08What is AI
- 8:02Challenges of AI
- 13:45AI ML in cyber security events
- 19:27How is AI affecting cyber security
- 21:09How AI affects cloud security
- 22:31Ransomware example
- 25:33Behavioral AI example
- 32:45AI potential for threat intelligence
- 36:38AI for security operations
- 38:03Summary
related talks
20mCatch Attackers Before They Strike with AI-powered Threat Intelligence
44mAI Meets Cybersecurity: Boosting Speed Without Losing Control
13mUsing AI to thwart cloud ransomware
57mAI as the New Hacker Frontier
16mIntelligent by design: Can cybersecurity cope with AI?
30mThe Simple, Yet Lethal, Anatomy of an AI Attack
20mSecurity, and AI... so happy together?
11mAI-Powered Cybersecurity in the Age of Advanced Threats