This article is really underrated; I think you perfectly captured both the strengths and weaknesses of AI search and the big question at the heart of it: who’s willing to pay for it?
I don't think Perplexity is going to survive. They haven't reached 10 million downloads on Android yet, whereas Openai as 100 million plus downloads. Real time search is an adjacent use case of a chatbot. I wrote an article recently explaining why your product should not look like a chatbot. Furthermore, build a product that can't be replicated inside a chatbot interface.
In a market where two competitors are building a relatively similar product, the company with superior distribution wins.
Perplexity is a chatbot with search. Openai has added search. Distribution wins.
Check out the article to understand the remaining 3 strategies
This article is really underrated; I think you perfectly captured both the strengths and weaknesses of AI search and the big question at the heart of it: who’s willing to pay for it?
I don't think Perplexity is going to survive. They haven't reached 10 million downloads on Android yet, whereas Openai as 100 million plus downloads. Real time search is an adjacent use case of a chatbot. I wrote an article recently explaining why your product should not look like a chatbot. Furthermore, build a product that can't be replicated inside a chatbot interface.
In a market where two competitors are building a relatively similar product, the company with superior distribution wins.
Perplexity is a chatbot with search. Openai has added search. Distribution wins.
Check out the article to understand the remaining 3 strategies
https://theriseofai.substack.com/p/ai-is-eating-software-how-to-build
Excellent article! Search and AI is a very nuanced- much more than people make it out to be. This is especially the case for monetization.
The one thing I don’t understand about SearchGPT is why it exists. Shouldn’t these features just be built into and piloted via regular ChatGPT?