TheChek News Flow

When OpenAI in March released ChatGPT features that could turn photos into animated characters, a spike in usage forced the company to put temporary limits on the features.

It wasn’t the first time this had happened. Chief Financial Officer Sarah Friar told investors at a Goldman Sachs conference last week that the company has to regularly hold up launches of new features or AI models because of a shortage of compute capacity, or it purposely has to make some of its products work more slowly. OpenAI is “massively compute constrained,” she said.

The Takeaway

One solution: the company plans to spend about $100 billion renting backup servers from cloud providers over the next five years, executives have told some shareholders. That’s on top of the $350 billion the company has already projected to spend on server rentals from cloud providers this year through 2030.

The backup plan reflects CEO Sam Altman’s belief that access to servers will be a defining advantage in the AI race; he has told colleagues OpenAI will someday need more power than the entire U.S. power grid to run its servers.

He and his executives believe future product hits or artificial intelligence breakthroughs will cause unforeseen surges in OpenAI’s computing needs, according to several people who have spoken to them.

Public comments from Altman and Friar imply OpenAI is already using all the large clusters of Nvidia graphics processing units it has access to. If it has another viral hit that strains its computing capacity or doesn’t have enough servers to train a promising new AI model, it would be at risk of losing more business or users to competitors such as Google or Meta Platforms that may not face the same problems.

Including the backup servers, OpenAI is planning to spend about $85 billion a year on server rentals over the next five years, on average, according to the company’s private projections. To put that into perspective, customers of Amazon Web Services, Microsoft, Google Cloud and Oracle collectively spent roughly $200 billion renting servers from the four firms last year.

OpenAI’s $100 billion in backup server costs is one reason why its long-term cash burn projections total $115 billion through 2029, The Information earlier reported, and why it may need to raise that much capital. The company also projected it will start generating cash in 2030.

Despite the extra costs, there is a silver lining in the backup servers. OpenAI executives say the servers are “monetizable” because they will help the company gain additional revenues it hasn’t yet included in its projections, either by powering research breakthroughs or a surge of usage of its products, according to a person who spoke to the executives. OpenAI has projected its revenue will rise to roughly $200 billion in 2030 from $13 billion this year, largely on the strength of ChatGPT. That revenue would be greater than what Nvidia or Meta Platforms generated in the past 12 months.

Hoarding Concerns?

OpenAI could also lower its back-up server costs or monetize the servers in other ways, including by giving back the excess capacity to Oracle, Microsoft or other cloud firms that provided it—but only if the cloud providers agree. Or it could find new customers for the server clusters, though it isn’t clear which companies would be in a position to spend that kind of money (other than perhaps OpenAI’s top competitors) and whether the cloud providers would allow OpenAI to resell the capacity.

If OpenAI doesn’t use the excess servers or can’t find other customers for them, it could be seen as hoarding GPUs. Nvidia, whose hardware currently powers virtually all of OpenAI’s servers, would likely want to avoid that result because it could deprive other companies of the chips and prompt them to use rival AI chips.

Executives have projected that OpenAI’s spending on the back-up servers will peak at $40 billion in 2028. It isn’t clear why the projections show a marked decline in those costs in later years, but OpenAI has separately said it will eventually develop its own data centers and chips, which could make up the difference.

“Think of OpenAI becoming very much a full-stack company,” Friar said at the Goldman Sachs conference.

Silicon Valley founders and executives are familiar with viral products that cause server shortages—and their potential consequences. For instance, outages involving Friendster, an early social network that took off more than 20 years ago, opened the door to competitors.

OpenAI experienced several outages in the past year or so, sometimes lasting multiple hours, though it managed to maintain its dominance in the chatbot market. The company is closing in on 1 billion weekly active users by the end of the year, after hitting 700 million weekly active users in July.

Ever since OpenAI acquired a hardware startup co-founded by former Apple design chief Jony Ive, the AI company has been ramping up its recruiting of people who work on Apple devices. It’s even tapping the iPhone maker’s supply chain network.

As OpenAI’s plans to make devices infused with its artificial intelligence get closer to reality, the company is increasingly plucking talent from Silicon Valley’s most successful device maker: Apple.

In recent months, the startup behind ChatGPT has been raiding the ranks of Apple’s design, manufacturing and supply chain teams, in addition to Apple’s team of AI researchers, according to ten current and former Apple employees and a review of LinkedIn profiles. OpenAI has also begun tapping the same supply chain network in China that Apple spent decades developing to help OpenAI with its efforts to make a new line of devices.

Luxshare, a major assembler of iPhones and AirPods in China, has already secured a contract to assemble at least one of OpenAI’s devices, according to multiple people with direct knowledge of the matter. OpenAI has also approached Goertek, which assembles AirPods, HomePods and Apple Watches, to supply components such as speaker modules for OpenAI’s future products, the people said.

One of the products OpenAI has talked to suppliers about making resembles a smart speaker without a display, the people said. OpenAI has also considered building glasses, a digital voice recorder and a wearable pin, and is targeting late 2026 or early 2027 for the release of its first devices, one of the people said.

One reason for OpenAI’s success in poaching Apple staffers is the lucrative compensation packages it has dangled in front of them. But the company isn’t just using money to win them over. Tang Tan—OpenAI’s chief hardware officer, who previously spent 25 years at Apple working on design—has promised some of the people he is recruiting from Apple that they will encounter less bureaucracy and more collaboration at OpenAI than they do at their current employer, said one person familiar with Tang’s outreach to Apple employees.

To say OpenAI still has a lot to prove in hardware would be a dramatic understatement. A long line of companies—from powerhouses like Amazon to scrappy young startups—have tried unsuccessfully for years to challenge Apple’s dominance in hardware with products like the iPhone and iPad. And novel devices built from the ground up around AI so far haven’t sold well either.

“Just because you’ve built something at a big company with billions of dollars of resources and the top people in the world doesn’t make you more qualified to do it outside,” said Matt Rogers, a former Apple engineer who worked on the development of the iPod and iPhone before co-founding smart home startup Nest. “A lot of my former Apple colleagues went off to start their own things and almost all of them were total failures.”

One prominent example: Humane, which recruited a large number of Apple hardware and operations experts, to help the startup make a wearable, AI-powered pin. The product launched in 2024 to poor reviews. Earlier this year, HP acquired Humane for $116 million and discontinued sales of the product.

Still, OpenAI has shown it’s willing to make major bets on hardware—and tapping the expertise of people who helped make Apple the most successful device maker is a big part of them. In May, OpenAI announced it was acquiring io Products—a hardware startup co-founded by former Apple design chief Jony Ive and Tan, and backed by OpenAI CEO Sam Altman—for $6.5 billion.

And if AI is destined to become the defining feature of devices in the future, as some technologists predict, OpenAI could have a key advantage over Apple and others thanks to its leading position in the category. The stakes for Apple, which gets over 70% of its revenue from device sales, are huge.

“Our goal was, and is, to create a family of products that would revolutionize the way people interact with AI technology,” Tan stated in a court declaration filed in a lawsuit against io Products and OpenAI over trademark infringement.

OpenAI and Apple declined to comment.

Less Incrementalism

OpenAI’s hiring spree from Apple has kicked into a higher gear since OpenAI announced the io deal.

So far this year, OpenAI has recruited more than two dozen employees from Apple who worked on consumer hardware, up from around 10 employees last year and virtually none in 2023, according to an analysis of LinkedIn profiles. Those recruits include hardware engineers and designers who previously focused on user interfaces, wearables, cameras and audio engineering at Apple, among other areas, according to their LinkedIn profiles and people familiar with the matter.

Among the new hires is Cyrus Daniel Irani, who joined OpenAI in July after a nearly 15-year career at Apple. Irani, most recently a director on Apple’s human interface design team, worked with Ive on projects such as the multicolored waveform that animated Siri’s voice response beginning with the iPhone 6S in 2015, according to a review of patent filings.

Other recent recruits to Tan’s team include Matt Theobald, a nearly 17-year Apple veteran working on manufacturing design, and Erik de Jong, a high-up on the Apple Watch hardware team, according to people familiar with their job moves (neither has updated their LinkedIn profile to indicate they work at OpenAI).

In some cases, OpenAI has successfully wooed Apple employees with offers that are significantly higher than their compensation at the iPhone maker, in large part because of OpenAI stock grants that can exceed $1 million, one person said.

OpenAI isn’t the only one initiating the recruiting. The news of the io deal also prompted an influx of Apple employees reaching out on their own to OpenAI about working there, according to one person familiar with the team’s recruiting.

A big part of the attraction of working at OpenAI for Apple staffers is the chance to reunite with Tan and Ive, according to former Apple employees.

A former head of product design for the iPhone and Apple Watch, Tan left Apple to co-found io last year. He previously reported to Apple’s head of hardware, John Ternus, and was widely respected within Apple for helping take the designs dreamed up by Ive’s team and transforming them into products that the company could manufacture at scale.

In some cases, Tan doesn’t appear to have to twist the arms of his former Apple colleagues to get them to join OpenAI.

Some longtime Apple employees working on the company’s hardware products have become bored with the incremental changes in the type of products they’re working on and frustrated with bureaucracy at Apple, said people familiar with their thinking. It hasn’t helped that employees have seen their incomes suffer due to Apple’s lackluster stock over the past year, the people said. Tan has told people that his vision is to re-create how industrial designers and hardware teams used to work together more efficiently and on bolder products at Apple, said a person familiar with Tang’s poaching discussions.

Ive’s involvement with OpenAI, too, has enhanced the company’s credibility in the eyes of its recruits from Apple. The designer played a critical role in the resurrection of Apple under Steve Jobs in the late 1990s and early 2000s.

His stature was sufficient that he was able to negotiate an agreement with OpenAI that allows him to continue to be involved with his own design firm, LoveFrom. The firm will continue to do work for various clients, which in the past have included everyone from Ferrari to Airbnb. LoveFrom, too, has poached talent from Apple, but the firm is relatively small and has mostly focused on hiring designers, not experts in hardware engineering, manufacturing and supply chain, The Information previously reported.

In recent months, OpenAI’s devices project seems to be consuming more of Ive’s time. He is often seen visiting the company’s San Francisco office, according to a person who has witnessed the visits.

OpenAI’s device ambitions and poaching from Apple could complicate the relationship between the two companies. Since 2024, the two companies have partnered to integrate OpenAI’s models into Apple’s voice assistant Siri and the image-generation app Image Playground. The two are currently discussing a deeper partnership through which OpenAI’s models would help power an overhauled Siri, said people familiar with the conversations.

OpenAI’s decision to rely on Chinese suppliers could also put the company in an awkward position. OpenAI’s policy teams have been openly critical of the Chinese Communist Party’s AI policies and called attention to alleged ties between China’s military and homegrown startups like Zhipu AI. Apple, in contrast, is extremely deferential in its rhetoric and actions related to China, a country that accounts for 16% of its revenues and where the majority of its products are manufactured.

There are some signs that Apple has become concerned about OpenAI’s recruiting of its employees. Last month, Apple abruptly canceled an offsite meeting it holds annually in China for some members of its U.S. and China manufacturing and supply chain teams, where senior managers typically review plans for future products with employees, according to a person with direct knowledge of the matter.

Its reason for ditching the meeting: Senior Apple leaders were concerned that the meeting would keep too many executives away from the company’s Cupertino, Calif., headquarters for too long, at a time when they were needed to prevent further defections to OpenAI, said a person who was informed of the decision.

Gapping down
In reaction to earnings/guidance:

Other news:

Gapping up
In reaction to earnings/guidance:

Other news: