Seemingly every company is obsessed with artificial intelligence these days, whether it’s how the technology is transforming their industry or the effects it’s having on employees and customers.

But the degree to which companies are utilizing AI tools internally and adapting to a rapidly changing reality varies dramatically. A new study from AI-Driven Enterprise Institute (AIDE) breaks down how well S&P 500 companies — and their leaders — are adopting AI compared to their peers.

The top performers, unsurprisingly, are centered in the tech industry, according to the data, which was shared with CNBC. In looking at four areas — literacy, advocacy, orientation and implementation — AIDE gave each company a score of up to 100 in the four categories and then provided an overall index score.

In technology, the highest company score (the average of the orientation and implementation pillars) and the only 100, went to chipmaker Nvidia, which has become the world’s largest company by selling the chips and systems that have powered the development of AI models and services. Meta and Amazon also scored 100, but in the S&P 500, those companies are considered communication services and consumer discretionary names, respectively.

The only other 100 company score went to energy producer SLB, formerly Schlumberger. The next highest scorer was retailer Walmart, followed by AES and NextEra Energy, which are both classified as utilities.

The new open-source index draws from publicly available data like earnings call transcripts, job openings and patent applications to measure how much executives know and say about AI, as well as how much their companies are prioritizing the technology and bringing it into daily operations.

The data doesn’t measure whether AI is driving financial returns, but it’s meant to give leaders an objective way to compare their strategy to their peers without relying on self-reported surveys, said Paul Cheek, AIDE’s CEO and a senior lecturer at Massachusetts Institute of Technology.

“When a board asks a CEO — ‘How are we doing compared to our peer group?’ — I don’t want it to be speculative,” Cheek said in an interview. “I want there to be some data that they can use to back up what they have to share.”

Cheek said there’s “significant room for improvement” for board members and executives to increase their own AI literacy, adding that boards need to better understand AI “as it relates to the ability to manage risk and strategic investments in the organizations that create value for all of us.”

Here are the the 20 companies with the top overall scores, based on their “orientation” and “implementation” scores:

1. Nvidia (100)
2. Schlumberger (100)
3. Amazon (100)
4. Meta (100)
5. Walmart (95.84)
6. AES (95.46)
7. NextEra Energy (95.44)
8. Ecolab (95)
9. Digital Realty (94.74)
10. Chevron (94.74)
11. Alphabet (94.72)
12. Equinix (94.59)
13. IQVIA (93.75)
14. Dow (93.34)
15. Halliburton (92.83)
16. Broadridge Financial Solutions (91.66)
17. Microsoft (91.37)
18. Block (90.91)
19. Duke Energy (90.91)
20. PepsiCo (90.62)

These companies were at the top of their sector based on the “AIDE Index”:

WATCH: Meta reshapes workforce as AI disrupts entry level hiring

Waymo is starting to offer select customers rides in its new Ojai robotaxis, which are designed to work better in snowier conditions and are cheaper to manufacture than the company’s older fleet of cars.

Rides will be available to some public passengers in San Francisco, Los Angeles and Phoenix, before opening up in San Diego, Las Vegas and Denver this summer, Waymo said in a statement on Thursday.

Alphabet’s self-driving division already had 100 Ojai vehicles on the road as part of its almost 4,000-car fleet. The company will have thousands of Ojai cars on the road by the end of the year, Ryan Powell, Waymo’s head of design, told CNBC in an interview. 

Waymo first announced the Ojai design in 2024. In February, the company opened rides to employees and their guests in San Francisco and Los Angeles.

As Waymo expands its fleet in an effort to cement its lead in the U.S. over Tesla and Amazon’s Zoox, the company is aiming to control costs. The Ojai is the first vehicle to take advantage of the sixth-generation Waymo Driver, the company’s proprietary driving system, which includes technology that can better detect critical details in low-light situations.

The vehicles, integrating new hardware and software, are made by China’s Geely, and cost significantly less to manufacture than earlier Waymo cars made by Jaguar and others. The Ojai robotaxis require fewer costly cameras and sensors than the previous Jaguar I-PACE vehicles.

Waymo, which got a $16 billion cash infusion in February from parent company Alphabet and other backers, offers fully autonomous robotaxis in several U.S. markets and plans to begin service in London later this year. The company has driven over 20 million overall autonomous rides and is looking to hit 1 million weekly trips by the end of the year, Powell said.

In the new vehicles, Waymo uses its own custom chips. The cars have also has been outfitted with improved lidar that can see through heavy rain and snow conditions, and have upgraded audio receivers that can better hear sirens and and other noises, the company said.

Ojai cars are roomier and also have removable steering wheels in them. The doors open and close automatically.

Powell said Waymo has learned from the design team that the “familiar form factor of the I-PACE and having the steering wheel and the pedals up there” helps people get “over that hump of trying something new.”

When asked whether the cars will integrate Google’s Gemini technology, Powell said the company is exploring more “natural ways of interacting with the vehicle” using AI technology. He gave the example of telling the car to move up a little further.

“I get excited about some of the things that you see on the Gemini side that may make some of those things possible,” he said.

While Waymo rapidly expands, the company is also dealing with some safety-related setbacks.

Earlier this month, Waymo issued a voluntary recall of about 3,800 of its robotaxis to fix software issues that could allow them to drive into flooded roadways, after incidents were captured on social media. Waymo said last week that it’s temporarily pausing freeway rides after spotting problems with its performance around construction zones. 

—CNBC’s Lora Kolodny contributed to this report.

WATCH: Waymo says it’s raised $16 billion at $126 billion valuation

Nvidia stepped on the gas last year, putting cash into companies up and down the AI infrastructure stack and helping to fund businesses that could turn around and buy the chipmaker’s technology. It’s been a lucrative endeavor, as the company’s $5 billion bet on Intel is now worth over $25 billion, a historic return in a matter of months.

In 2026, the pace of deals has kicked into overdrive, with Nvidia already topping $40 billion in commitments and expanding its portfolio to include more public equities.

This week alone Nvidia forged an agreement with data center operator IREN, giving it the right to invest up to $2.1 billion in the company, a day after Nvidia struck a pact with Corning, allowing it to invest up to $3.2 billion in the 175-year-old glass maker. Shares of IREN and Corning popped on the announcements.

Nvidia has been the biggest winner of the artificial intelligence boom, producing the graphics processing units required to train AI models and run large workloads. The global scramble to secure GPUs has lifted Nvidia’s stock by more than 11-fold in four years, propelling the company to a roughly $5.2 trillion market cap and making it the most valuable business in the world.

To help the company grow its dominance beyond chips, Nvidia is financing the entire AI supply chain, ensuring it runs on Nvidia hardware and that there’s sufficient capacity to meet demand. But there’s growing concern in some corners of AI that Nvidia — like cloud providers Google and Amazon — is investing in other companies as a way to fuel its own growth.

Nvidia, which generated $97 billion in free cash flow last fiscal year, is backing some of the very companies that buy its chips and, in some cases, is leasing compute right back to them. Critics have compared it to the vendor financing that helped inflate the dot-com bubble.

Matthew Bryson, an analyst at Wedbush Securities, said in a note that Nvidia’s investments and buildouts fit “squarely into the circular investment theme” that’s been driving fears around the market’s durability. However, Bryson sees the investments as underscoring Nvidia’s vision and creating a “competitive moat” if the company can execute.

An Nvidia spokesperson didn’t respond to a request for comment.

Nvidia has signed at least seven multibillion-dollar investments this year with publicly traded companies. Additionally, it’s been part of roughly two dozen investment rounds in private companies, including some relatively early-stage deals, according to FactSet.

During the last fiscal year, Nvidia invested $17.5 billion in private companies and infrastructure funds, “primarily to support early‑stage startups,” according to its annual filing with the SEC. The company said those investments include AI model companies that purchase its products directly or through cloud service providers.

Non-marketable equity securities (which are private company investments) held on Nvidia’s balance sheet swelled to $22.25 billion at the end of January from $3.39 billion a year earlier. The company reported gains on those assets as well as publicly held equities of $8.92 billion, up from $1.03 billion in the prior fiscal year, in part due its investment in Intel, which has turned into a stock market darling this year, up well over 200%.

On Nvidia’s last earnings call in February, Huang said, “Our investments are focused very squarely, strategically on expanding and deepening our ecosystem reach.”

The IREN deal this week includes an agreement that the data center company will deploy up to 5 gigawatts of Nvidia’s DSX-branded infrastructure designs intended to power AI workloads at facilities across the globe.

As part of the Corning deal, the glass company is building three new U.S. facilities dedicated to optical technologies for Nvidia, which will likely be turning to fiber-optic cables instead of copper as it builds out its rack-scale systems.

In March, Nvidia invested $2 billion in Marvell Technology as part of a strategic partnership to work on silicon photonics technology. That month it put the same amount in Lumentum and Coherent, two companies developing photonics technologies.

Then there are the so-called neoclouds. In January, Nvidia invested $2 billion in CoreWeave in a deal that involves building out data centers with Nvidia’s technology. It also invested $2 billion in Nebius Group, an AI cloud company, as part of an agreement on AI infrastructure deployment, fleet management, inference and AI factory design.

Chip analyst Jordan Klein at Mizuho called the deals with component makers “super smart by the CFO and team and a great use of cash,” because they help accelerate the development of critical technology and products that are in short supply. He’s more skeptical of the neocloud investments, which he said “feel more questionable to me and likely investors.”

“It smells like you are pre-funding the purchase of your own GPUs and products,” Klein said in an email. Still, he noted that the cloud providers have critical attributes like power and data center capacity that Nvidia needs.

Ben Bajarin at Creative Strategies shared a similar sentiment regarding IREN, telling CNBC, “The risk is that if the cycle turns, the market starts questioning how much of the demand was organic versus supported by Nvidia’s own balance sheet.”

As much as Nvidia is funneling money into publicly traded partners, those wagers are dwarfed by the chipmaker’s investment in OpenAI.

The $30 billion Nvidia pumped into OpenAI in late February came more than a decade after the companies started working together, though they’ve grown increasingly intertwined since the launch of ChatGPT in 2022, the moment that sparked the generative AI frenzy.

Nvidia’s investment in OpenAI was originally going to be much bigger. In September, the companies said Nvidia would be putting in up to $100 billion over time into OpenAI as the AI company deployed 10 gigawatt’s of Nvidia’s systems. That deal never got off the ground as OpenAI pivoted away from developing data centers, instead leaning heavily on partners like Oracle, Microsoft and Amazon to piece together as much capacity as possible.

Huang said in March that investing $100 billion in OpenAI is probably “not in the cards,” and that the $30 billion deal “might be the last time” it writes a check before an IPO that could take place this year.

WATCH: Nvidia’s AI supply chain empire: Here’s what you need to know

Shares of companies tied to artificial intelligence infrastructure fell Tuesday after a report that OpenAI has fallen short of internal growth expectations, raising fresh questions about whether the pace of spending across the sector is sustainable.

Oracle, which has a $300 billion, five-year partnership to supply computing power to OpenAI for AI operations, dropped 4%.

Chipmakers including Broadcom and Advanced Micro Devices declined 4% and 3%, respectively. Shares of Nvidia fell more than 1%.

Qualcomm dipped 0.2% Tuesday but closed off its lows. The stock had gotten a slight boost Monday on reports it is working with OpenAI on smartphone chips tied to the firm’s hardware ambitions. Leveraged neocloud stock CoreWeave dropped more than 5%.

In Asia, SoftBank Group, one of OpenAI’s largest investors, sank about 10%.

The Wall Street Journal reported that OpenAI has recently missed its own projections for user growth and revenue. The shortfall has sparked internal concern about whether the company can keep pace with the massive financial commitments required to build out data centers and secure long-term computing capacity.

According to the report, finance chief Sarah Friar has warned colleagues that if revenue growth doesn’t accelerate, the company could face difficulty funding future compute agreements.

OpenAI pushed back on the report. “This is ridiculous. We are totally aligned on buying as much compute as we can and working hard on it together every day,” the company told CNBC.

Oracle defended OpenAI’s growth trajectory, saying it’s seeing firsthand how quickly adoption of OpenAI’s technology is accelerating.

“We’re incredibly excited about our partnership with OpenAI and remain focused on building and delivering the capacity they need to support rapidly growing demand,” an Oracle spokesperson said. “OpenAI’s new 5.5 model is a significant step forward, and we expect continued momentum as access to their technology expands across cloud providers.”

Open AI, which kickstarted the AI boom with the launch of its ChatGPT chatbot in 2022, recently closed a record-breaking $122 billion funding round at a post-money valuation of $852 billion.

“You would assume any slowing was known by the investors, right? If not, shame on OAI,” Jordan Klein, TMT sector specialist at Mizuho, said in a note. “How new could update be as the round closed end March when the quarter would have ended. And it’s not even May 1. I highly doubt OAI fundamentals slowed that fast in under 30 days.”

Meanwhile, competition in enterprise AI is intensifying. Anthropic has been gaining traction with corporate customers, while Google’s Gemini models are also picking up momentum as companies increasingly adopt multiple providers.

Still, some investors questioned whether the report materially changes the broader AI spending outlook.

“I view the article as largely a rehash of what we already knew: OpenAI’s growth seems to have slowed in late-2025 into early-2026 as the business ceded some share to Anthropic and Gemini,” said John Belton, portfolio manager at Gabelli Funds. “There is nothing here that suggests this is an issue for the pace of spending across the sector as a whole; instead, this looks more like confirmation about OpenAI’s recent market share trends.”

Others emphasized how difficult it is to apply traditional metrics to the current AI cycle. Luke Rahbari, CEO of Equity Armor Investments, said shortfalls in revenue targets should be viewed with caution, given how imprecise forecasting remains in a rapidly evolving industry.

“OpenAI missing its revenue targets is, in the grand scheme, a distraction,” said Rahbari. “In the current AI landscape, these projections are largely arbitrary. No major player in this race can accurately forecast their revenue or capital expenditure within a 25% to 50% margin of error.”

— CNBC’s Kate Rooney and Seema Mody contributed reporting.

Chinese artificial intelligence startup DeepSeek on Friday released a preview version of its long-awaited V4 large language model, allowing users to test its new capabilities and features. 

The release comes more than a year after the Hangzhou-based company introduced its R1 reasoning model, which rocked global tech markets due to its surprising performance and cost efficiency.

Similar to DeepSeek’s previous model releases, the latest upgrade is open-source, allowing developers to download the code, run it locally and modify it in most cases.

The model is available in both a “pro” and a “flash” version, depending on size, with DeepSeek claiming that V4 achieves strong performance against domestic competitors, particularly in agent-based tasks, knowledge processing and inference.

“DeepSeek’s V4 preview is a serious flex,” offering lower inference costs than previous models, Neil Shah, vice president of research at Counterpoint Research, told CNBC.

Inference costs refer to the computational and financial expenses of running a trained AI model to generate outputs.

DeepSeek also said that V4 has been optimized for use with popular agent tools such as Anthropic’s Claude Code and OpenClaw.

According to Counterpoint’s principal AI analyst, Wei Sun, V4’s benchmark profile suggests it could offer “excellent agent capability at significantly lower cost.”

Elon Musk is seeking to have OpenAI CEO Sam Altman and President Greg Brockman removed from their roles as officers in the company as part of a case that’s expected to go to trial later this month.

In a legal filing on Tuesday, Musk’s lawyers laid out specific remedies their client is seeking if a judge and jury determine that Altman and OpenAI defrauded Musk, the world’s richest person.

Musk sued Altman and OpenAI in 2024, claiming the artificial intelligence company that he helped start almost a decade earlier “assiduously manipulated” and “deceived” him into donating $38 million, based on promises that the entity would remain a nonprofit. The two sides have since been embroiled in a public war of words, in addition to their legal battle and budding business rivalry.

“Plaintiff will seek an order removing Altman as a director from the OpenAI nonprofit board and removing both Altman and Brockman as officers of the OpenAI for-profit,” Musk’s lawyers said in Tuesday’s filing. “Removal of a charity’s officers and directors is a common remedy where those individuals fail to protect or carry out the charity’s public mission.”

Musk is also asking the court to make OpenAI revert to operating as an actual nonprofit, according to the filing. The company completed a restructuring in October, and is now run as a nonprofit with a 26% stake in the for-profit arm, which includes ChatGPT.

Jury selection for the case is slated to begin on April 27, in a federal court in Oakland, California.

Following Tuesday’s filing, OpenAI said in a post on X that Musk is “pretending to change his tune about attacking the nonprofit OpenAI Foundation.”

“The truth is that this case has always been about Elon generating more power and more money for what he wants,” OpenAI said. “His lawsuit remains nothing more than a harassment campaign that’s driven by ego, jealousy and a desire to slow down a competitor.”

Musk, Altman and others co-founded OpenAI as a nonprofit AI lab in 2015. Musk left OpenAI in 2018, after trying to convince executives there to merge it with Tesla, his electric vehicle company.

In 2023, Musk launched a competing company called xAI, which developed the AI image generator and chatbot Grok. In February, Musk’s SpaceX acquired xAI, which also owns X (formerly Twitter) in a deal that valued the combined entity at $1.25 trillion. SpaceX recently filed confidential paperwork with the SEC for what will likely be a record IPO.

On Monday, OpenAI sent a letter to the California and Delaware attorneys general, urging them to investigate “improper and anti-competitive behavior” by Musk and his associates ahead of the trial. In the letter, OpenAI strategy chief Jason Kwon alleged that Musk has been working to undermine OpenAI through various “attacks” on the company, including by “coordinating his efforts” with Meta CEO Mark Zuckerberg.

Musk’s attorneys previously said, in a January filing, that their client should receive up to $134 billion in damages from OpenAI ⁠and lead investor Microsoft, calling them “wrongful gains” that the companies had received as a result of his early work with and financial support of OpenAI.

In Tuesday’s filing, Musk’s lawyers said their client is seeking “to return all ill-gotten gains, including Microsoft’s, to the OpenAI charity.”

— CNBC’s Ashley Capoot contributed to this report.

WATCH: OpenAI is chasing Anthropic

A version of this article first appeared in CNBC’s Inside Wealth newsletter with Robert Frank, a weekly guide to the high-net-worth investor and consumer. Sign up to receive future editions, straight to your inbox.

Investment firms of ultra-wealthy families dialed back their deal-making in March as the Iran conflict rattled the market.

Family offices made 39 direct investments in companies last month, a 25% drop from February when adjusted for month length, according to data provided exclusively to CNBC by Fintrx, a private wealth intelligence platform.

That said, the family offices that are still inking deals are making bold bets. A quarter of last month’s investments were part of mega-rounds, or fundraises in excess of $100 million, according to Fintrx.

In March, Jeff Bezos’ namesake family office co-led a $1.03 billion seed round for Advanced Machine Intelligence. Also known as AMI Labs, the new startup is training artificial intelligence models on real-world sensory data, rather than text.

Other boldface-name billionaires such as ex-Google CEO Eric Schmidt and serial entrepreneur Mark Cuban also participated in the fundraise.

This trend of making fewer but larger deals is also playing out with corporate investors.

This past quarter, the total value of global mergers and acquisitions activity rose by 26% compared with the same quarter last year to $1.2 trillion, but the number of deals fell by 17%, according to data from LSEG. The second week of March was the worst week for global M&A in over a year, falling below $33 billion, LSEG found.

However, some family offices continue to be prolific dealmakers.

In March, Indian billionaire Azim Premji’s family office made at least four direct investments in companies, according to Fintrx. Premji Invest’s largest round, which it also led, was a $450 million Series A for Rhoda AI, another startup developing novel ways to train artificial intelligence models. Rhoda AI aims to train industrial robots on hundreds of millions of videos. Kleiner Perkins billionaire John Doerr also backed the round.

Elon Musk’s legal team is demanding that a judge in the Delaware Court of Chancery, Kathaleen McCormick, recuse herself from two Tesla lawsuits, alleging she showed bias in supporting a LinkedIn post that was critical of Musk.

“In light of the Court’s recent public support of LinkedIn posts that create a perception of bias against Mr. Musk in these cases, recusal is necessary and warranted,” Musk’s attorney’s wrote in their motion for recusal on Wednesday. “These cases should be re-assigned to another random-drawn judicial officer of this Court.”

The post that McCormick allegedly responded to with an emoji had touted a verdict from a San Francisco federal court that could cost Musk upwards of $2 billion. In that case, a jury found Musk had defrauded Twitter investors in the lead up to his buyout of the social network.

McCormick sent a letter to Musk’s attorneys saying she did not read the full text of the LinkedIn post referenced, did not support it, or intend to click any emoji expressing support of the post, and would have only accidentally clicked that indicator. She also said in the letter that she had reported “suspicious activity” on her account to LinkedIn.

Quinn Emanuel partner Alex Spiro, an attorney for Musk, told CNBC in an email that he’s inquired with LinkedIn on the matter, and said “I don’t believe there’s any basis for the claim it was a ‘glitch.'”

LinkedIn, which is owned by Microsoft, did not immediately respond to a request for comment.

McCormick became the target of Musk’s ire after she ordered Tesla to rescind his 2018 CEO pay package, worth about $56 billion in options, when she presided over the shareholder suit Tornetta v. Musk.

Musk moved his businesses, including Tesla, out of Delaware, incorporating them in Texas and Nevada and encouraging others to do the same.

In 2025, Delaware’s Supreme Court said Musk’s 2018 pay package must be restored, deciding that the lower court’s decision by McCormick was too extreme a remedy and did not give Tesla a chance to say what a fair compensation for Musk ought to be.

Tesla and Musk still have two cases proceeding through Delaware courts before McCormick. One concerns Tesla directors’ compensation, and the other is a consolidated shareholder suit filed by investors alleging that Musk breached his fiduciary duties to Tesla when he started a potential competitor in artificial intelligence, xAI,

Earlier this year, Tesla invested $2 billion into xAI. Musk’s aerospace and defense company, SpaceX, then acquired xAI, converting Tesla’s stock into SpaceX holdings.

WATCH: Elon Musk teases pair of giant chip factories

In a document that resembles an IPO prospectus, OpenAI said its close ties with Microsoft could be a potential risk to its business, telling investors that the software company is responsible for “a substantial portion of our financing and compute.”

OpenAI included sections titled “Risks Related to the Transaction” and “Risks Related to our Business” in a financial document, viewed by CNBC, that the company shared with prospective investors tied to its recent record financing round.

Last month, OpenAI announced $110 billion in funding from strategic partners including Amazon, Nvidia, and SoftBank. The company is working with banking partners to tack on an additional $10 billion worth of commitments from a broader pool of investors, according to sources familiar with the deal. That part of the round is on track to close by the end March, said the people, who asked not to be named because the details are confidential.

The risks highlighted by OpenAI offer a taste of what’s to come in its upcoming IPO filing, as the company gears up to make its public market debut as soon as this year. Aside from its relationship with Microsoft, OpenAI cited risks such as its significant capital expenditures, reliance on compute resources, ongoing litigation with Elon Musk’s xAI, and its unusual structure as a public benefit corporation, whose parent is the OpenAI Foundation.

OpenAI was founded as a nonprofit research lab in 2015, but has experienced exploding commercial growth since launching ChatGPT to the public in late 2022. ChatGPT now boasts 900 million weekly active users, and the company generated $13.1 billion in 2025 revenue. It was valued last month at $730 billion by investors.

Microsoft has been a backer since 2019, years before ChatGPT was released, and obtained an early commitment from OpenAI to move some of its services exclusively to Microsoft’s Azure cloud. In total, Microsoft has invested $13 billion in OpenAI and, at the time of the AI company’s restructuring in October, disclosed that its 27% diluted stake in the for-profit part of the organization was valued at $135 billion.

OpenAI said in the document circulated to investors that its operating results will depend on its ability to successfully develop relationships with additional partners aside from Microsoft.

“If Microsoft modifies or terminates its commercial partnership with us, or if we are unable to successfully diversify our business partners, our business, prospects, operating results and financial condition could be adversely affected,” the company wrote.

An OpenAI spokesperson said in a statement that, “This is a standard legal risk factor disclosure, unrelated to any potential IPO prospectus.”

“Similar language has been in place for years,” the spokesperson said. “Microsoft is and will remain a critical long term partner.”

Though OpenAI and Microsoft have a tight bond, they’re increasingly competing for users in the burgeoning generative AI market.

In 2024, Microsoft added OpenAI to the list of competitors in its annual report, a roster that for years has included Amazon, Apple, Google and Meta. And last year, OpenAI turned to other cloud providers, such as CoreWeave, Google and Oracle, to meet heavy demand.

Geopolitical, legal risks

While Microsoft warranted its own headline in the risk disclosures, it’s not the only company that OpenAI named.

OpenAI noted that it requires enormous amounts of computational resources to train and run its AI models, and that a global chip shortage could be damaging.

Specifically, OpenAI said that if chip supplier Taiwan Semiconductor Manufacturing Company is affected by a regional conflict, a nod to the growing tensions between China and Taiwan, OpenAI could face “severe disruptions” to its supply chain.

OpenAI also said it expects to continue making significant capital expenditures and commitments for “compute, data center services and related infrastructure projects,” alongside partners Microsoft, Nvidia, Advanced Micro Devices and Broadcom.

As of December, OpenAI said it had roughly $665 billion in estimated compute spend commitments through 2030, according to the document, adding that “our compute requirements are dynamic and may expand significantly.”

Then there’s the long and expanding list of legal cases.

OpenAI cautioned investors that ongoing litigation could be problematic due to copyright, patent and other intellectual property issues, along with employment and contract disputes, privacy concerns and other matters.

The company detailed three different lawsuits filed by OpenAI co-founder Musk or his company, xAI, which is now part of SpaceX after a merger last month. Musk left OpenAI in 2018, after trying to convince executives to merge it with Tesla. The two sides have been involved in legal battles since 2024, with the first case expected to go to trial next month.

OpenAI also said that at least 14 lawsuits have been filed against the company in California state and federal courts by ChatGPT users or their family members, who blame the company’s products for “mental illness leading to suicide, death or other injury.”

The first wrongful death lawsuit was filed in California last year by Matt and Maria Raine, parents of 16-year-old Adam Raine, who died by suicide after ChatGPT reportedly encouraged him to take his own life. 

“We are reviewing these cases, in light of our existing industry-leading safeguards and additional efforts, as well as the complex nature of the causes of mental illness,” OpenAI said in the document.

One name is notably absent from OpenAI’s risk factors section: Sam Altman.

The CEO and co-founder has long been the public face of the company and has been mired in his share of controversy. In late 2023, Altman was suddenly ousted by the board, which said it has lost confidence in the leader, but then reinstated him days later due to employee and investor pressure.

OpenAI acknowledged in the document that the “success of our company and the operation of our business rely on Key Personnel.” Neither Altman nor any of his colleagues are identified.

If you are having suicidal thoughts or are in distress, contact the Suicide & Crisis Lifeline at 988 for support and assistance from a trained counselor.

WATCH: OpenAI renews focus on enterprise in all-hands meeting amid IPO push