ChatGPT & Search? NeevaAI’s Real-Time AI Search Is Already Here - Ep. 300 via @sejournal, @lorenbaker

It seems like everyone these days wants to talk about artificial intelligence (AI) and its potential in search.

It’s true – there is no doubt that AI will continue revolutionizing how we find information online. But did you know that real-time AI search is already here? Introducing NeevaAI, it harnesses the power of artificial intelligence for more efficient search results in real time.

Neeva CEO and ex-Head of Google Ads, Sridhar Ramaswamy, joined me on the SEJ Show to discuss how live AI questions and answers will change how people search. We touched on how AI will alter the business models of search engines. And also covered the limitations of ChatGPT, and the importance of sourcing and citations in AI-powered live search-driven answers.

We felt that commercial search engines, Google, and Bing, had become products for advertisers, not products for you and me to find out what we want. We felt that a user-focused search engine could differentiate itself and deliver a better product experience. So Neeva, at its core, is a user-first search engine. That means that we don’t show ads, don’t show affiliate links, and we protect your privacy. –Sridhar Ramaswamy, 02:03

NeevaAI seemed to fix most of the issues I was experiencing with chatGPT, especially the lack of sourcing and accurate information.
–Loren Baker, 35:30

The underlying technology behind chatGPT is this model called GPT-3.5. The core technique that that model uses is learning on every page, every piece of text that it can get its hands on. Now note they don’t know where these pages came from. They don’t know whether it is real or not. They don’t understand things like link structure authority. –Sridhar Ramaswamy, 06:41

[00:00] – About Sridhar.
[05:29] – Limitation of ChatGPT.
[10:17] – A walkthrough on Neeva’s AI search.
[17:35] – Monetization challenges ad-supported search engines may face with AI.
[21:24] – How verified sources are served.
[23:48] – How personalized are AI-driven answers based on history?
[24:54] – Will there be more follow-up queries after the AI results are given?
[27:43] – Where this is going in terms of privacy.
[35:30] – Is Neeva available in multiple languages?
[37:12] – Can website owners verify with Neeva, like Google Search Console?

Resources mentioned:
Neeva: ad-free private search: https://neeva.com/

All of us, willingly or not, are participating in a giant experiment similar to Google. We are happily submitting our thoughts, worries, and desires to chatGPT. And worse, it is not equipped to compartmentalize information like Google reasonably can. Google’s like, okay, this is your account, these are your searches, you want to delete it, tell me. We’ll delete it. They make it hard for you to do it, but it is possible. So I think there needs to be oversight over how this is done. What storing personal information is going to mean? If the data is then used for training future models, that’s when life gets very scary.–Sridhar Ramaswamy, 32:50

One of the main points of concern that I’ve had with some peers while discussing AI is real-time answers. This could be a direct negative on the traditional Google search monetization model.–Loren Baker, 15:59

Lack of authority is a huge issue. Also, these models are humongous. They take an enormous amount of money to train and an enormous amount of money to run, and they’re not real-time. So what we do at Neeva is we use a technique called retrieval, augmented generation. That’s a fancy way of saying to use the power of search to set the reality field for a large language model. Then, ask it to generate text and cite which part of the input text it uses.–Sridhar Ramaswamy, 08:27

For more content like this, subscribe to our YouTube channel: https://www.youtube.com/user/searchenginejournal

Connect with Sridhar Ramaswamy

Sridhar Ramaswamy is an inspiring example of using technology for the greater good. After leading Google’s enormous advertising division, he succeeded as a venture capitalist at Greylock Partners.

He then took his ambitions one step further with Neeva. This privacy-focused search engine has no ads and runs on a subscription-based model instead.

This tech master combines humanistic values with revolutionary thinking to make the software work towards beneficial societal outcomes. With such ambition inspiring him forward, this entrepreneur has set down new paths of discovery. As a result, we can all benefit from Sridhar’s tireless vision for innovation!

Connect with Sridhar on LinkedIn: https://www.linkedin.com/in/nancy-marzouk-82519a/
Follow him on Twitter: https://twitter.com/RamaswmySridhar

Connect with Loren Baker, Founder of Search Engine Journal:

Follow him on Twitter: https://www.twitter.com/lorenbaker
Connect with him on LinkedIn: https://www.linkedin.com/in/lorenbaker