Innovation & Tech

Bill Gross Wants Fortune 500 Firms to Pay $250,000 a Year for AI Patent Mining

Bill Gross is licensing his AI idea-generation system to Fortune 500 firms at $250,000 a year, after a Caltech test turned 9,978 patents into 3,867 business leads.

By Nathan Brooks

4 min read

‘What if the input isn’t the news? What if the input is you?’ Bill Gross gave away 1.2 million AI business ideas for fre
‘What if the input isn’t the news? What if the input is you?’ Bill Gross gave away 1.2 million AI business ideas for freblickpixel / pixabay

What's News

  • Bill Gross plans to license his AI idea system to corporations at $250,000 a year, with five $25,000 one-year pilots.
  • A Caltech test ran 9,978 patents and produced 3,867 businesses Caltech should build or license.
  • The public Idealab.ai system generated 1.2 million ideas in about a month for roughly $1 million in AI costs.
  • 131,000 people across 202 countries have searched the free idea database since its October launch.
  • Research matching 1.4 million products to patents found big firms' patents tie to launches far less than smaller firms'.

Idealab founder Bill Gross wants Fortune 500 companies to pay $250,000 a year to run his AI system on their own patents — a month after he gave away 1.2 million AI-generated business ideas for free.

The pitch landed Monday at the Fortune 500 Innovation Forum in Detroit, in a session sponsored by Idealab.ai. "What if the input isn't the news? What if the input is you?" Gross told the executives in the room. He meant their patents, trade secrets and, in his words, "all the things that you've shelved inside of your company."

The commercial structure is already on the table. Gross said Idealab plans to license the full system to companies at $250,000 a year, and it's offering five companies a one-year pilot for $25,000. The corporate revenue, he argued, is what keeps the public version free: "Big companies with 100 billion dollar year R and D budgets will pay for it, and that way we can give it away free to entrepreneurs."

What problem is Gross actually selling against?

Big companies hold enormous patent portfolios that never become products. That's not a hunch — research from Kellogg's Sara Moreira, with economists at Yale and the Atlanta Fed, matched about 1.4 million products to patents and found that large firms' patents are tied to new product launches "far less so" than smaller firms'.

Moreira's explanation cuts against Gross's framing, though. Big companies often use patents as a "defensive tool" to block competitors rather than to launch anything. In plain terms: some of that shelved R&D sits on the shelf on purpose.

The Caltech test

Gross launched Idealab.ai in October after feeding an AI system a sprawling dataset: news stories going back to 2000, every patent since 1980, product recalls, and complaints from Reddit and X, combined with his own 31 methods for generating business ideas. The system produced 1.2 million ideas in roughly a month, at about $1 million in AI compute costs. Since launch, he said, 131,000 people across 202 countries have searched them.

To test the corporate version, Gross went to Caltech — his alma mater, where he's a trustee and, as he noted on stage, a board member. He asked the tech-transfer office for four patents.

"They said, 'Why just four? We'll give you all 9,000,'" he recounted. The system ran all 9,978 Caltech patents and generated 3,867 businesses that Caltech "should either build or license," Gross said.

Put that in context. Caltech's tech-transfer office signed 59 licenses and options and helped form eight startups in fiscal 2024, per its annual report. The AI system flagged 3,867 opportunities in a single run.

The teaser before the sale

Idealab didn't wait for the forum to make its case. It ran the public patents of the 113 attending companies through the system and printed each one a sheet of business ideas "that you're not already doing," Gross said.

"Now that's a teaser," he added. "What we really want to do is work with you and have you give us the things that aren't public." In other words, the printed sheets are the appetizer; the real product requires access to trade secrets and unpublished IP.

Gross has a long historical frame for all this. He said he modeled Idealab, founded in 1996, on Thomas Edison's lab, and quoted Edison on stage: "I find out what the world needs, then I go ahead and try to invent it."

What does this change for corporate R&D?

For practitioners, the offer is simple: a $25,000 pilot buys a machine-generated audit of your patent portfolio's commercial potential — work that tech-transfer offices like Caltech's do at a pace of dozens of licenses a year, not thousands.

The catch is the research. If a chunk of your patents exists to fence off competitors rather than to ship products, a system that flags 3,867 "opportunities" may be surfacing decisions your company already made deliberately. The bet Gross is asking executives to make is that their shelf is bigger than their strategy intended — and now there's a number on it.

Source: fortune.com

Share this article:

More from Nathan Brooks

Nathan Brooks

Show full bio

News editor covering regulation, policy and their impact on entrepreneurship, startups and venture funding at Business Bearings.

2,352 articles

Related articles

« Previous articleNext article »