What’s more, he says Pearl is much less likely to provide false information than many other AI-powered search engines, which he says will likely cope with a “wave” of lawsuits based on the wrong answers they provide. “These other players are creating amazing technologies. I call them Ferraris or Lamborghinis,” says Kurtzig. “We are building Volvo – safety comes first.”
Of course, this comment about Pearl’s superiority made me want to try it even more. Kurtzig seemed so confident that Pearl would continue to enjoy protection under Section 230. I asked AI if he agreed.
Pearl stated that it likely qualifies as an “interactive computer service” under Section 230, meaning it would not be treated as a publisher, as Kurtzig suspected. But AI added: “Pearl’s situation is unique because it generates content using AI.” In the end, it didn’t give me a definitive answer.
When I asked to speak to a lawyer directly, I was directed to JustAnswer, where I was asked to provide the answer I wanted to verify. I said I had to go back and copy the response because it was several paragraphs long, but when I went back to Pearl’s site, the conversation had disappeared and was reset to a new chat.
When I tried again, this time opening the Pearl browser on my computer, I received a similarly shaky response. I decided to run a human fact check; within minutes I had a TrustScore™ – a measly 3!
Pearl directed me to get a real expert opinion by taking me to the subscription page. I received a login so I didn’t have to pay while testing the tool. He then put me in touch with one of his “legal eagle” experts.
Unfortunately, the lawyer’s answers were no clearer than the AI’s. He noted that there is an ongoing legal debate about the application of Section 230 to AI search engines and other AI tools, but when I asked him to make specific arguments, he gave an odd answer, noting that “most employ shell companies or associations to submit applications.” “
When I asked for an example of one of these shell companies – quite confused as to what this had to do with the public debate over Section 230 – the “legal eagle” asked if I wanted him to put together a package. Even more confused, I replied yes. I got a pop-up saying that my expert wanted to charge me an additional $165 to look up the information.
I refused, frustrated.
I then asked Pearl about the WIRED story. The AI answer was useful, although basically the same stuff you’d find on Wikipedia. When I asked for TrustScore™, I got a 3 again, which suggested it wasn’t a very good answer. I chose to connect with another human expert. This time, perhaps because it was a media question rather than a plain legal or medical topic, it took some time for the expert to appear – well over 20 minutes. When he did this, the expert (what gave him his media bona fides was never determined, although his profile indicated he had been with JustAnswer since 2010) gave me an eerily similar answer to AI. Since I was running a free test, it didn’t matter, but I would be annoyed if I actually paid a subscription fee just to get the same mediocre response from both a human and an AI.
The last time I used the service, I asked a plain question: how to renovate the kitchen floor. This time everything went much calmer. The AI returned an appropriate response that resembled a transcription of a very basic YouTube tutorial. When I asked the expert to assign a TrustScore™, he received a 5. The score certainly seemed precise enough. But – as someone who really wants to refinish the elderly pine boards in my kitchen myself – I think when I actually go looking for tips, I’ll rely on other online communities of human voices that don’t charge $28 a month: YouTube and Reddit.
If you end up testing Pearl or another cutting-edge AI search product and have a memorable experience, let me know how it went in the comments below. You can also contact me by email at kate_knibbs@wired.com. Thanks for reading and stay protected!