Perplexity AI CEO Believes No Publisher Should Have Rights to Facts

December 7, 2022, seven days after OpenAI ChatGPT was launched by a former OpenAI research scientist named Aravind Srinivaswho left the company just three months earlier, launched a competing artificial intelligence chatbot called Perplexity. “Everyone was obsessed with ChatGPT. We were the only product that came and said: links and citations are important. So from the very beginning we took care of this.” Srinivas, Co-founder and CEO AI confusionhe said during an on-stage interview at TechCrunch Disrupt on Oct. 30.

Perplexity has been in hot water lately for the very problem Srinivas set out to solve two years ago. Last month, the company was sued by News Corp.’s Wall Street Journal and New York Post for plagiarizing their content in search results. A few days earlier, the New York Times sent “cease and desist” a notice to a startup requiring it to stop using the newspaper’s content on its website.

Perplexity is at the forefront of so-called artificial intelligence answering systems, which aim to answer users’ specific questions by summarizing information online, rather than simply providing links in response to a few keywords. Srinivas said the average query entered into Perplexity’s answering system is 10 to 11 words, compared with two to three words in Google Search, suggesting users are coming to Perplexity with more well-thought-out questions.

Srinivas said Perplexity “always cites its sources” and “does not claim ownership of any content.” “It’s just bringing content from the Internet, summarizing it in a way that the user can digest it, and then telling you where they’re getting all that information from,” he said, adding that it’s the same way journalists do their jobs. and therefore should not be considered plagiarism.

However, he acknowledged that, as with other rapidly evolving AI applications, Perplexity’s current guardrails are not perfect and can be easily bypassed through rapid design, a buzzword that describes the practice of designing inputs to AI tools that will produce optimal results. .

The new publications suing Perplexity claim the AI ​​company is competing for the same audience as them by using copyrighted content. But Srinivas said Perplexity users don’t log into the app to read the daily news, but to “make sense of what’s going on.”

“Like, how will this particular news affect me? In the context of the news, should I continue to buy more Nvidia shares? These are not questions you can ask TechCrunch, but you can come and ask Perplexity,” the CEO said.

Recognizing that the news being communicated is important to enhancing the value of Perplexity’s product, earlier this year the company launched a unique program share advertising revenue with news publishers. He currently contributes to Time, Fortune and the German news site Der Spiegel.

But ultimately, Srinivas believes that no one should have a right to the facts. “We believe that the facts must be communicated to everyone,” he said. “Imagine a world in which scientists claim ownership of a certain fact, but other people cannot claim it. Knowledge and truth cannot be spread in this way.”

Srinivas, originally from India, holds a Ph.D. He received a bachelor’s degree in computer science from the University of California, Berkeley, and was a research intern at DeepMind and Google. Before co-founding Perplexity, he worked as a research scientist at OpenAI for about a year.