By the way, for a while, even when you were working on the Forbes article, it linked you to the Business Insider article. So the “rough edges” are right. Again, I have no problem. I’m an AI bull. I have no problem with people trying to figure it out, but maybe don’t introduce it until you figure it out. Or maybe we should also figure out what the model is so that all sites benefit, because this completely challenges the idea of how publishers and search engines have worked for the last 15 years.
Lauren Goode: And as far as I understand, Forbes has actually prevented AI-powered search engines like Perplexity from browsing its site by essentially disabling robot.txt’s search engine crawlers. And yet somehow it still happens?
Randall Lane: It raises many larger issues. I think the more vital question is: how can we coexist? Again, Google, which has a trillion-dollar market capitalization, is probably worth as much as almost all the publishers in the world combined. However, there is a relationship and partnership where Google sends a lot of traffic to many publishers and makes an incredible amount of money from it. But how will it work when these kinds of summarizing machines basically publish?
So they call themselves a search engine again, but they act like a publisher, and that’s a one-way ticket. And I think that’s the tipping point that we have here, where technology, I think people, it’s one of those things where people say, “Well, this is a problem that falls on its head.” Problems Here. So what do we do about it?
Kate Knibbs: What media outlets like Forbes can do to prevent this, apart from speaking out? Maybe Perplexity will really, really take this criticism to heart and completely change its model. But if that doesn’t happen, where do we go from here?
Randall Lane: Well, there are many paths. There could be a legal route, which we don’t want, but when people steal, it’s obvious and easiest. But in the long run you become a game of whack-a-mole. It’s not very practical. Secondly, there is constant war. But the third path, with generally accepted technology, is: what is coexistence that is fair to all parties?
You see, OpenAI has started working with publishers, but it’s a trained partnership. This is different from taking content and creating derivative content, which largely replicates the base content. So we are entering a up-to-date sphere. Embarrassingly, he quote, unquote, “coincidentally” right after this incident said, “Well, we’re implementing a new partnership model with publishers.” But again, it remains to be seen what this will look like. Taking something that costs us money, literally stories like this, when you’re talking about all the reporting and all the time that goes into graphics and design and editing, you’re talking about content that costs thousands or, even in this case, tens of thousands of dollars.