Bill Gross Wants To Save The Internet From AI.

Bill Gross has been here before. 

Back when the Internet was young, when the dot-com wave broke across the monied shoals of Wall Street opportunism, Gross built a world-changing company, took it public, then sold it for billions. 

Google was a fledgling startup with no discernible business model when Gross pioneered a revolutionary approach to Internet advertising now known as “paid search.” It seems obvious today, but in 1998, the notion that advertisers could buy search keywords and only pay if people clicked on their ad - well, that idea was bonkers. 

So bonkers, in fact, that when Gross launched Goto.com at TED in February of 1998, the audience booed and hissed while he was on stage. The gall of this man! This bespeckled, wiry, manic guy from Los Angeles of all places - he wanted to ruin the Internet by putting a price tag on search results!(1) 

Despite the initial response from the TED crowd, Goto flourished, growing through the dotcom crash that devastated most of the early Internet industry. Knowing he could never catch the portals dominating early Internet traffic, Gross pivoted his company to a syndication model, executed a successful IPO, and changed Goto’s name to Overture. By 2002, the company had search monetization partnerships with Microsoft, AOL, and Yahoo - the three largest Internet destinations of the era.

Gross had the right idea at the right time - an idea that Google went on to borrow, improve, and capitalize upon; an idea that drove a profound shift not only in how businesses operate, but also in how our society values, processes, and governs its most fundamental currencies: Information and human attention.(2)

Bill Gross got rich from that idea - he sold his company to Yahoo, his largest customer, in 2003. But if you ask Gross whether things turned out the way he had hoped they would, his answer is a resounding no. Gross surveys the landscape of today’s Internet, and once again, he sees a massive opportunity hiding behind an equally massive problem: Everyone’s using these new AI engines, but so far, no one’s figured out how to make them pay. 

***

Search built the modern Internet. Google organized the world’s information and made it accessible, in the process producing a massive fuel depot that powers yet another revolution: Generative AI. AI’s gluttonous maw - large language models from OpenAI, Google, Meta, Amazon, Claude, and a thousand startups - consume today’s Internet like a pack of ravenous ouroboroi, ripping through the fabric of established business models like “paid search” in the process.  

So of course Gross wants back in, to once again produce an idea so bonkers it could get him booed off the stage at TED(3). But the world has changed dramatically since he last stepped forward as the CEO of an audacious startup. Can Bill Gross do it again? 

ProRata.AI is his answer, the 150th (or so) company Gross has launched through Idealab, a Wonka-esque incubator he founded nearly thirty years ago. Every ten or so years Idealab starts a company so important, so potentially game-changing, that Gross decides to take the role of CEO. ProRata is that kind of company. 

The Internet today is broken, Gross argues. The grand bargain between Internet sites and search engines - we’ll let you crawl our site if you send us traffic in return - that bargain has been subverted by monopolistic Internet giants who refuse to share the wealth, as well as craven startups eager to cash in on the AI boom. And now that billions of consumers have shifted their attention towards generative AI chatbots, the “rest of the web” - the very seed corn that feeds those bots - is withering. 

It’s not that Gross thinks answer engines are bad - quite the opposite. Services like ChatGPT are a “10X improvement” on search, he tells me. “They’re incredible,” he says, “but they need a new monetization model.” 

The model that he built for traditional search won’t work for AI-driven answer engines, Gross explains. And that’s where ProRata comes in. The company’s audacious idea? To build a monetization engine that does for generative AI what Overture did for search some 25 years ago. 

***

Gross needs to solve two problems if ProRata’s vision is to become a reality. First, he has to catalyze a marketplace for publishers’ content - a way for publishers to get paid when AI crawlers ingest and regurgitate content. And secondly, he has to build an advertising solution that works in the context of an answer engine - essentially, he has to figure out what the ads should look like in a world dominated by AI. 

No big deal, right? 

Gross already has help with the first problem. Those ravenous AI bots hoovering up websites at a rate of thousands of crawls a day? They’re shoplifting, Gross says. AI services should pay for the privilege of ransacking the open Internet, he argues. This concept - “pay per crawl” - has already taken root: Internet infrastructure giant CloudFlare has implemented a pay-per-crawl marketplace premised on a similar philosophy. Publishers that aren’t being paid by those data-hungry AI bots can now avail themselves of a free service from CloudFlare that blocks them at the door. 

Now….imagine if all websites blocked AI agents! How might the world change?! Well, those AI services would have no choice but to start paying site owners. But that raises our second problem: AI services need a revenue stream that they could share with publishers - one that scales well beyond the subscription models which currently dominate services like Claude and ChatGPT. “The AI companies are all complaining that they can't afford to pay,” Gross tells me. “Well, if I can triple their revenue, then they can afford to pay, right?”

Advertising has always been the lifeblood of the Internet. With ProRata, Gross has built an advertising solution that generates ads on the fly based on the context of each users’ queries. To catalyze his marketplace, ProRata operates an AI answer engine called Gist.AI that features content from licensed partners who are paid each time the Gist service uses the partner’s content to generate an answer. Gist also serves as a proof of concept for ProRata’s advertising solution, which splits revenues with its publishing partners. Here’s what it looks like:

That ad for the EV in the middle of the page? That’s ProRata’s revenue model in action.

Gross knows that Gist.ai will never get to the scale necessary to compete with the ChatGPTs of the world. So just as he did with Overture, Gross is leaning into a syndication model - distributing Gist “answer bots” across publisher sites, signing up advertisers, and working to sign monetization deals with the AI answer engines of the world. 

If Gross has his way, a significant number of both publishers and AI chatbots will avail themselves of ProRata’s services, creating a marketplace similar in nature to the one Goto and Overture created 25 years ago.(4) 

I was fortunate enough to pay attention to Gross and when he launched Goto back in 1998 - I later devoted a chapter in The Search to his story. Gross doesn’t have to sign Google Gemini or OpenAI to succeed - if he garners even one-tenth of the current AI answer engine market, he’ll have built a company far bigger than Overture, and potentially just as impactful. 

Will ProRata prove to be Gross's second act? I’d not bet against it - and if ProRata's model takes root, it will mark a significant departure from how the Internet has worked to date.  More than two decades after sparking a revolution in search, Bill Gross is once again driving a controversial and counter-intuitive startup, one designed to equitably monetize a massive shift in technology, capital, and consumer behavior. Keep an eye on ProRata. I smell a good story brewing. 

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.

1. Back in the day, TED was a very different kind of conference. Its audience had a strong anti-commercial vibe - pitching from the stage was seen as declassé, and every so often a speaker would endure boos or hisses if their talk offended the community’s sensibilities. I happened to attend those early TEDs, and I launched The Industry Standard at the same 1998 gathering where Gross was booed. In the room were not only Gross and I, but Jeff Bezos, who has a cameo in Goto’s launch narrative, as well as Sergey Brin and Larry Page of Google. 
2. Google launched the same year as Goto - but with a decidedly anti-advertising philosophy. “We expect that advertising funded search engines will be inherently biased towards the advertisers and away from the needs of the consumers,” wrote Google founders Sergey Brin and Larry Page in a research paper that launched Google. “the better the search engine is, the fewer advertisements will be needed for the consumer to find what they want.” As Google grew and became a consumer destination in its own right, Gross held talks with Brin and Page, pitching either a merger or a commercial partnership. Instead, Google’s founders decided to emulate and improve on Gross’s model.
3. Alas, TED only lets people give standing ovations these days….
4. Of course, there’s one big difference: The Internet market today is orders of magnitude larger than it was in 2003.