"The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis

The Pixel Revolution Part 2 with Suhail Doshi, Founder of Playground AI

Jan 26, 2024
Suhail Doshi, Founder and CEO of Playground AI, dives into the intricate world of AI image generation and manipulation. He shares insights on the progress and hurdles in technology, contrasting the rapid evolution of language models with the slower pace of vision advancements. They explore innovative techniques like outpainting and inpainting, and discuss the ethical implications of generative AI. Suhail also highlights the exciting transition from static images to dynamic video forms in the art world, emphasizing the role of AI in fostering creative expression.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Limited Utility of Image Models

  • AI image models excel at creating art but lack broader utility like language models.
  • Vision models are currently at the GPT-2 stage, with limited practical applications beyond art.
INSIGHT

Image Models at GPT-2 Level

  • Current image models are like GPT-2, good for a few tasks but lack general utility.
  • They excel at art but struggle with logos, spatial reasoning, and detailed editing.
INSIGHT

Lack of Imagination Limiting Image Model Utility

  • The limited utility of current image models is mainly due to a lack of imagination.
  • People struggle to envision the potential of these models beyond art creation.
Get the Snipd Podcast app to discover more snips from this episode
Get the app