Can this AI-powered search engine replace Google? It has for me.

Throughout my adult life, whenever I had a question about the world or needed to look something up online, I turned to Google for answers.

But recently I took to Google with a new AI-powered search engine. (No, not Bing, who is dead to me after trying to break up my marriage last year.)

It's called Bewilderment. The birthday boy search engine, whose founders previously worked in AI research at OpenAI and Meta, has quickly become one of the most talked-about products in the tech world. Tech insiders am excited about it on social media, and investors like Jeff Bezos – who was also an early investor in Google – have showered it with cash. Company recently announced that it had raised $74 million in a funding round led by Institutional Venture Partners, which valued the company at $520 million.

Many startups have tried to challenge Google over the years but have failed. (One potential competitor, Neeva, shut down last year after failing to gain traction.) But Google seems less invincible these days. Many users have complained that their Google search results have become clogged with spammy and low-quality websites some people have started searching for answers on places like Reddit and TikTok.

Intrigued by the hype, I recently spent several weeks using Perplexity as my default search engine on both desktop and mobile. I tested both the free version and the paid product Perplexity Pro, which costs $20 per month and gives users access to more powerful AI models and certain features, such as the ability to upload their own files.

Hundreds of searches later, I can report that while Perplexity isn't perfect, it is very good. And while I'm not ready to break up with Google completely, I'm now more convinced that AI-powered search engines like Perplexity can loosen Google's grip on the search market, or at least force it to catch up to make.

I also worry that AI search engines could destroy my job, and that the entire digital media industry could collapse because of products like these. But I'm getting ahead of myself.

Where it shines

At first glance, Perplexity's desktop interface looks a lot like Google's: a text box centered on a sparse landing page.

But once you start typing, the differences become clear. When you ask a question, Perplexity doesn't return a list of links. Instead, it scours the internet for you and uses AI to write a summary of what it finds. These answers are annotated with links to the sources the AI ​​used, which also appear in a panel above the answer.

I tested Perplexity on hundreds of questions, including questions about current events (“How did Nikki Haley do in the New Hampshire primary?”), shopping recommendations (“What's the best dog food for an older dog with joint pain?”), and household questions . tasks (“How long does beef stew last in the refrigerator?”).

Each time I got back an AI-generated response, usually one or two paragraphs long, sprinkled with quotes to websites like NPR, The New York Times, and Reddit, along with a list of suggested follow-up questions I could ask, such as “Can you make beef stew freeze it so it lasts longer?”

An impressive Perplexity feature is 'Copilot', which allows a user to refine a search by asking clarifying questions. For example, when I asked for ideas on where to host a birthday party for a 2-year-old, Copilot asked if I wanted suggestions for outdoor spaces, indoor spaces, or both. When I selected “inside,” I was asked to choose a rough budget for the party. Only then did I get a list of possible locations.

Perplexity also allows users to search within a specific set of sources, such as academic articles, YouTube videos, or Reddit posts. This came in handy when I was looking for how to change a setting on my home's water heater. (Exciting stuff, I know.) A Google search turned up a number of unhelpful links to DIY tutorials, some of which were thinly veiled advertisements for plumbing companies. I tried the same search on Perplexity and limited my search to YouTube videos. Perplexity found the video I needed for my exact water heater model, extracted the relevant information from the video and converted it into step-by-step instructions.

Under the hood, Perplexity runs on OpenAI's GPT-3.5 model along with its own AI model – a variant of Meta's open-source Llama 2 model. Users who upgrade to the Pro version can choose from a handful of different models, including GPT-4 and Claude from Anthropic. (I used GPT-4 for most of my searches, but I didn't see much difference in the quality of the answers when I chose other models.)

Bewilderment is also refreshingly good at admitting it when it occurs not know something. Sometimes a partial answer was given to my question, with a caveat like “No further details are provided in the search results.” Most AI chat products I've used lack this kind of humility: their responses sound confident even when they're spouting nonsense.

Where Google still rules

In my testing, I found Perplexity most useful for complex or open-ended searches, such as summarizing recent news articles about a specific company or providing suggestions for date-night restaurants. I also found it useful when what I was looking for—instructions for renewing a passport, for example—was buried on a busy, difficult-to-navigate website.

But I did sneak back to Google for a few types of searches, usually when I was looking up specific people or trying to go to websites I already knew existed. For example, when I typed “Wayback Machine” into my browser's search bar, I was redirected to Perplexity, which spit out a one-paragraph essay on the history of the Internet Archive, the organization that maintains the Wayback Machine. I had to look for a little citation link to get to the Wayback Machine website, which is what I wanted in the first place.

Something similar happened when I asked Perplexity for directions to a work meeting. Google would have given me turn-by-turn directions from my home, thanks to its integration with Google Maps. But Perplexity doesn't know where I live, so the best it could offer me was a link to MapQuest.

Location data is just one of the many advantages Google has over Perplexity. Size is another: Perplexity, which has just 41 employees and is based in a shared workspace in San Francisco, has 10 million monthly active users, an impressive number for a young startup but a speck compared to Google's billions .

Bewilderment also lacks a lucrative business model. Currently, the site has no ads and fewer than 100,000 people pay for the premium version, says Aravind Srinivas, the company's CEO. (Mr. Srinivas didn't rule out a move to an ad-based model in the future.) And of course, Perplexity doesn't offer versions of Gmail, Google Chrome, Google Docs, or any of the dozens of other products that make Google's ecosystem so inescapable.

Mr. Srinivas told me in an interview that he believed Google was a formidable competitor, but that he thought a small, focused startup could spook the company.

“What gives me confidence is the fact that if they want to do better than us, they basically have to change their own business model,” he said.

What about hallucinations?

One problem with AI-based search engines is that they tend to hallucinate, or make up answers, and sometimes stray from their source material. This problem has dogged several AI search hybrids, including Google's first release from Bardand it remains one of the biggest obstacles to mass adoption.

During my testing, I found that Perplexity's answers were largely accurate – or, to be more precise, they were as accurate as the sources they relied on.

I did find a few errors. When I asked Perplexity when Novak Djokovic's next tennis match was, I was given the details of a match he had already played. Another time, when I uploaded a PDF file of a new AI research paper and asked Perplexity to summarize it, I got a summary of a completely different paper published three years ago.

Mr Srinivas acknowledged that AI-powered search engines still make mistakes. He said that because Perplexity was a small, relatively obscure product, users didn't expect it to be as authoritative as Google – and that Google would struggle to build generative AI into its search engine because it would lose its reputation for accuracy. had to uphold.

“Suppose you use our product and we do well on eight out of ten searches. You would be impressed,” Mr. Srinivas said. “Now suppose you use Google's product and it only gets seven out of ten points. You'd say, 'How can Google three questions wrong?'”

“That asymmetry is our opportunity,” he added.

A win for users, a loss for publishers

Even though I enjoyed using Perplexity, and I will likely continue to use it along with Google, I have to admit that I got a gnawing feeling in my stomach after seeing that it provided pristine, concise summaries of news stories, product reviews, and how-tos. to articles.

Much of today's digital media economy still relies on a steady stream of people clicking on Google links and seeing ads on publishers' websites.

But with Perplexity, there's usually no need to visit a website at all; the AI ​​does the browsing for you and gives you all the information you need, right on the answer page.

The possibility that AI-powered search engines could replace Google traffic – or push Google to put similar features into its search engine, as it has begun to do with its search engine “seek generative experience” experiment – ​​is part of the reason why many digital publishers are terrified right now. It's also one reason some are fighting back, including The Times, which last year sued OpenAI and Microsoft for copyright infringement.

After using Perplexity and hearing about it similar products are being developed Other start-ups have convinced me that the worriers have a point. If AI search engines can reliably summarize what's happening in Gaza, or tell users which toaster to buy, why would anyone ever visit a publisher's website again? Why would journalists, bloggers and product reviewers continue to put their work online if an AI search engine just swallows it up and regurgitates it?

I brought this fear to the attention of Mr. Srinivas, who responded with a diplomatic evasion. He admitted that Perplexity would likely send less traffic to websites than traditional search engines. But he said the traffic that remained would be higher quality and easier for publishers to monetize because it would be the result of better, more targeted searches.

I'm skeptical of that argument, and I'm still nervous about what the future holds for writers, publishers, and people who consume media online.

So for now, I'll have to weigh the convenience of using Perplexity against the concern that, by using it, I'm contributing to my own demise.