Nvidia paused its revenue-sharing deals with AI cloud firms over antitrust fears

Free daily briefing on global business news.

The chip giant stepped back from the AI Compute Partnership program last week, less than two months after announcing it in July

Cheng Xin / Getty Images

Nvidia $NVDA has stepped away from a financing program under which it provided credit backing to AI cloud providers in return for a cut of their revenue, according to The Wall Street Journal. Nvidia employees raised alarms with current and potential customers about the risk of antitrust exposure.

Nvidia pulled the plug on the initiative just last week, fewer than two months after announcing it in July. The Journal reported that Nvidia has left open the possibility of restructuring or absorbing the initiative into a different effort down the line.

"The new business model we introduced in July that opens up compute access to the fast-growing AI ecosystem is still in place and continues to evolve due to high demand," an Nvidia spokesperson said.

The program, called the AI Compute Partnership, was designed to address a financing problem facing smaller cloud providers, who must spend billions of dollars on Nvidia chips and data center infrastructure before securing enough customer contracts to borrow against. Under the arrangement, Nvidia promised to rent GPU capacity itself if a provider could not find another customer — giving providers a source of guaranteed revenue to support financing.

The structure gave Nvidia two ways to profit: revenue from chip sales and a portion of the income cloud providers earned from renting that capacity. Nvidia and each cloud provider would establish a base hourly rate covering the provider's costs, and Nvidia would receive 50% of any revenue above that threshold.

During the program's early weeks, Nvidia's insistence on maintaining significant control irritated a number of prospective partners. Providers were told they had to vet customers through an approval process, and Nvidia made clear it wanted utilization spread across multiple smaller AI firms instead of concentrated with one large buyer. A number of cloud providers objected, contending that client selection should remain their own prerogative.

A quarterly filing published this week marked the first time Nvidia disclosed the scale of the program, revealing $36 billion in total commitments tied to agreements with a typical six-year duration. CFO Colette Kress addressed the revenue-share program on a Wednesday investor call, noting that the new income source had the potential to bring in billions of dollars for Nvidia over the medium to long term.

The two companies initially named as partners in the program were Sharon AI and Firmus Technologies, the company said.

Join 500,000+ readers who start their day with Quartz.

By subscribing, you agree to our Terms of Service and Privacy Policy.

Global business news for a smarter world