Generative artificial intelligence search engine startup Perplexity AI said today that it has raised $73.6 million in new funding that places the company at a valuation of $520 million.

The company’s Series B funding round, led by Institutional Venture Partners LLC, was also joined by existing investors from its previous Series A and seed rounds, including NEA, former Twitter Vice President Elad Gil and Databricks Ventures. Also participating were new investors such as AI chipmaker giant Nvidia Corp., Bessemer Venture Partners, Factorial Funds, Kindred Ventures, Shopify Chief Executive Tobi Lutke and Amazon.com Inc. founder Jeff Bezos through the Bezos Expeditions Fund.

The news follows a report from The Information in October that the company was seeking funding and brings the total raised by the company to $100 million to date.

Perplexity’s search tools operate differently than a traditional search engine. Instead of requiring people to type in keywords that deliver a list of results, it allows users to pose questions and the search engine responds conversationally. In this way, it is similar to other AI chatbot search experiences such as Google Bard and Microsoft Corp.’s Bing Chat powered by OpenAI’s GPT-4.

“With Perplexity’s search tools, users get instant, reliable answers to any question with complete sources and citations included,” said co-founder and Chief Executive Aravind Srinivas. He explained that the company built its search chatbot to be accurate and transparent when providing information to users in an era when search engine optimization has made it difficult for users to sift through results. It also helps them avoid the issues of advertising and sponsored links, which clutter results and make it harder to understand what they’re looking at.

Generative AI allows the system to look through search results and turn them into conversational humanlike speech. They’re formed into understandable paragraphs of information, including citations, that can be consumed in a way that is more like holding a conversation than looking at the index or glossary of a book.

Under the hood, Perplexity is powered by multiple AI large language models, including models developed by the company in-house and third-party models such as Google’s Gemini, Mistra 7BI, Anthropic’s Claude 2.1 and OpenAI’s GPT-4. The choice of models to use is available to users who subscribe to the company’s Pro plan at $20 per month, which provides availability to Perplexity Copilot, which unlocks many features including personal preferences, file uploads and image generation.

Users also can choose between different “focus” modes for Copilot, which allows them to narrow the search domain. At its broadest, a user could have it search the entire internet, or it could be restricted to search only within scholarly publications.

It can also be set to “Wolfram Alpha,” which causes the AI to focus on computational and mathematical calculations. For community-focused searching, it’s possible to search only YouTube, which offers timestamps and videos, or Reddit, which helps sift through discussions on the forum site.

“Copilot tailors search queries with custom follow-up questions, introducing the concept of generative user interfaces,” said Srinivas. “It removes the burden of prompt engineering and does not require users to ask perfectly phrased questions to get the answers they seek. This enables users to gain more relevant and comprehensive answers than other AI chatbots, traditional search engines, or research tools.”

Users on the free plan can ask Copilot five questions every four hours, while subscribers get more than 600 searches a day.

Srinivas said that since the public launch of Perplexity a year ago, the conversational AI search engine app has grown to 10 million monthly active users and it has served more than a half-billion queries in 2023. During the same period, the mobile app has been installed more than a million times across iOS and Android.

He went on to say that he believes that the world is at a turning point when it comes to how people consume and search for information and that generative AI will be the mechanism that serves that roadmap.

 Image: Perplexity AI