Decoder with Nilay Patel We are not ready for better deepfakes
185 snips
Jul 24, 2025 
Gaurav Misra, CEO of Captions, dives into the world of deepfake technology, revealing its alarming advancements and potential dangers. He discusses the ethical implications of AI-generated videos and the rising concern over misinformation in political contexts. Misra emphasizes the urgent need for awareness, regulation, and innovative solutions to combat deepfake abuse, while exploring the current limitations of verification methods. With deepfakes becoming increasingly sophisticated, the conversation highlights our precarious balance between innovation and safeguarding authenticity in a digital age.
AI Snips
Chapters
Transcript
Episode notes
Diffusion Models Generate Whole Videos
- New diffusion-based models can generate entire videos from nothing, without original footage.
- This removes original references, making fake videos harder to identify and moderate.
Unlimited Length, Multi-Person Deepfakes
- Fourth generation deepfakes will have unlimited length and multi-person interaction with consistent settings.
- This transforms content creation into a software-like process, enabling entire movie scenes without physical sets.
Audio Integrity Lags Visuals
- Audio remains a key tell for advanced deepfakes due to lack of contextual background noise and casual tone.
- Synchronized audio-video models like Google's are emerging but still imperfect and undertrained.