AI-Generated Synthetic Media, aka Deepfakes and its impact
summary
Deepfakes are already practical: face swaps, lip‑sync, puppeteering, full image and voice cloning plus GPT‑style text generation create hyperreal falsifications used for revenge porn and political manipulation. Detection is weak — the Facebook challenge tops at 65% precision — and platforms see real harms like a European case where a cloned CEO voice led to a €250,000 transfer. Defenses blend content removal, dissemination limits (WhatsApp forwarding cap), provenance/authentication, labels, media literacy and policy/regulatory work to limit exposure and belief.
AI-written summary · about
chapters (15)
- 0:00Intro
- 1:05Information Disorder
- 1:56What is Deepfake
- 2:42Defects
- 3:42Types of Defects
- 5:49Positive Use Cases
- 9:09Harms
- 14:48Political Example
- 15:25Public Service Announcement
- 16:08Activism
- 16:44Technology
- 17:46Liars dividend
- 19:02Countermeasures
- 23:14Ethics
- 25:15Questions
related talks
30mReality Under Attack: Just How Serious Is the Deep Fake Threat?
36mDisinform your Surroundings: AI and disinformation campaigns
16mThe Threat of Deepfakes to Remote Identity Verification Systems
35mDisinform your Surroundings: AI and disinformation campaigns
27mAl-Generated Attacks: The New Era of Threats to Authentication
27mTruth or Fiction The Art of Unveiling DeepFakes in Text, Audio, Image
41mThe Deepfake Supply Chain: Weaponizing Synthetic Media into Modern Cybercrime Operation
26mDeepfakes, audiofakes, and the future of media