The world's top deepfake expert stopped trusting his own eyes. So our fix when nothing's verifiable: ask the AI that made the fake if it's real.

Deepfake video crossed the line where it can no longer be reliably detected, and the expert who would know lost the arms race himself. Hany Farid, the UC Berkeley digital forensics expert news desks call when nobody can tell whether footage is real, told the New York Times he can no longer distinguish a real video from AI. Hunter Powers and Daniel Bishop walk through the case at the center of it (a viral video, shared by an Iranian news agency and viewed over a million times, of a missile hitting an elementary school, that a full frame-by-frame analysis could not prove was fake) and how forensics teams used to catch AI video by detecting a real pulse in a person's face. From there: why your phone photos are already AI, image upscaling, the dead internet theory, and the show's actual fix, which is to ask the AI that may have made the video whether it is real, ideally after flipping it to maximum truthfulness mode. Then the money: Hunter's confession that he pays $300 a month for Grok Super Heavy, the full AI subscription stack (OpenAI, Anthropic's Claude, Google, Perplexity, Grok Imagine, Kling), why these plans keep getting more expensive, and whether running a frontier model like GLM locally at home (around $45,000 today) is where this goes. They land on trust and accountability: people now believe ChatGPT and Claude the way they believe the New York Times, the Munich court that threw out Google's defense of its AI Overview, Section 230, and the self-driving-car logic that ends with AI radiologists and AI lawyers becoming mandatory. When the thing lying to you has no pulse and no name, who do you hold accountable? They Might Be Self-Aware is the AI podcast from The Blur, reported from inside the dissolving line between human and machine, not from a safe distance.

CHAPTERS
0:00 Cold Open (Gary's Intro)
2:03 Deepfake Pulse Detection
4:05 Hany Farid Concedes
9:46 Just Ask the AI
14:42 Dead Internet Theory
17:52 Grok Super Heavy Plan
19:47 AI Subscription Stack
24:55 AI Trusted Like News
28:55 Google AI Overview Lawsuit
33:21 Self-Driving Car Logic

LISTEN / WATCH EVERYWHERE
🎧 Apple Podcasts: https://podcasts.apple.com/us/podcast/they-might-be-self-aware/id1730993297
🎧 Spotify: https://open.spotify.com/show/3EcvzkWDRFwnmIXoh7S4Mb?si=3d0f8920382649cc
🎧 Everywhere else plus episode page: https://theblur.ai

THE BLUR
Follow: @TheBlurAI

COMMENT
Hunter trusts the AI more than he trusts people now. Daniel will only go as far as "more than people, most of the time." Where do you actually land: trust the machine, or trust the human?

You're listening to They Might Be Self-Aware, from The Blur.
New episodes Monday and Thursday.

#Deepfake #AI #TMBSA #DeadInternetTheory