Unsupervised Learning Ep 23: Perplexity CEO Aravind Srinivas on the future of Search, OpenAI Wrappers and Using AI to Talk to Loved Ones
127 snips
Dec 14, 2023 
Aravind Srinivas, CEO and Co-Founder of Perplexity AI, talks about the birth and future of Perplexity, its potential as a Google competitor, and what search will look like in 10 years. They also discuss the challenges of being viewed as a 'wrapper' for OpenAI, the future of Quora and Wikipedia, and the potential regulation of AI. OpenAI's changing direction and the challenges of developing accurate search results are also discussed.
AI Snips
Chapters
Transcript
Episode notes
Build Models Later
- Start with existing models if product-focused, and build user base first.
- Build your own models later to control destiny and reduce dependency, like Perplexity AI.
Moats and Users
- Moats are relevant only when there's something substantial to protect.
- A "wrapper" with users is more valuable than a unique model with none.
Knowledge Velocity
- Perplexity AI aims to maximize knowledge velocity, unlike traditional knowledge platforms.
- It focuses on accelerating the rate of knowledge acquisition through personalization.