Tackling Deepfakes, Wars, and Other Security Threats in the GenAI World
summary
Panel details concrete threats and mitigations: hyperreal audio/video enables a $25M fraud via 15 wire transfers, text‑to‑video is live, and watermarking proves fragile. Agencies (CISA/FBI/DOD) and ABA task force push guidance while NIST SP800‑63 v4 promises updated identity controls. Defenses focus on liveness detection, revamping identity enrollment and proofing (resolution, validation, verification), metadata signals, multi‑factor blind‑spot hardening, and urgent cross‑sector information sharing.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 2:54Deepfakes and courtroom evidence
- 4:26Digital identity subversion
- 6:59Election interference risks
- 7:44Autonomous weapons misuse
- 8:30Deepfake detection challenges
- 15:06Deepfakes in courtrooms
- 20:14AI identity fraud risks
- 27:01Global election interference
- 29:10Military AI applications
- 32:06Solutions for Deepfakes
- 35:58ABA AI Task Force
- 41:10AI in legal practice
- 44:05Final recommendations
related talks
39mBeyond Recognition: A Cautionary Tale of AI and Identity Fraud
16mThe Threat of Deepfakes to Remote Identity Verification Systems
30mReality Under Attack: Just How Serious Is the Deep Fake Threat?
40mBanking Cybersecurity: Battling DeepFakes & AI-powered Scammers
15mDeep-sixing the deepfakes
41mDeepfake Threats in Business: What You Need to Know
28mTrust VS Tech – Fraud Prevention in the Era of Deepfakes and Voice Cloning
49mGenAI and Cybersecurity: Safeguarding the Digital Age