CoreWeave revenue beats estimates on AI boom but shares fall on bigger loss
The company currently operates 33 AI data centers across the U.S. and Europe and offers access to backer Nvidia's chips, which are highly coveted by enterprises to train and run large AI models amid intense competition.
CoreWeave reported revenue backlog of $30.1 billion as of end of June, compared with $25.9 billion on March 31.
"Demand is humming, but it is the cost of growth that tempered the stock down in aftermarket trading," said Michael Ashley Schulman of Running Point Capital Advisors.
Operating expenses jumped to $1.19 billion in the second quarter, from $317.7 million a year earlier.
The company posted a net loss of $290.5 million, compared with analysts' average estimate of $190.6 million, according to data compiled by LSEG.
"We are scaling rapidly as we look to meet the unprecedented demand for AI," CEO Michael Intrator said in the earnings statement.
Investors have also focused on the company's reliance on a few big customers.
"The backlog surge to $30B+ suggests demand visibility well beyond 2025, but the concentration in mega-customers like OpenAI means those relationships remain both the crown jewel and the single point of failure," said eMarketer analyst Jeremy Goldman.
Meanwhile, CoreWeave executives reiterated the benefits of its $9 billion all-stock deal for crypto miner Core Scientific.
The deal, announced in July, is facing opposition. Core Scientific's largest shareholder, Two Seas Capital, has said it would vote against the sale.
CoreWeave reported second-quarter revenue of $1.21 billion, beating estimates of $1.08 billion.
The Livingston, New Jersey-based company raised its annual revenue forecast to be between $5.15 billion and $5.35 billion. It had previously projected annual revenue of $4.9 billion to $5.1 billion.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
CNA
2 hours ago
- CNA
Analysis:Just in time? Manufacturers turn to AI to weather tariff storm
LONDON :Manufacturers like U.S. lawnmower maker The Toro Company are not panicking at the prospect of U.S. President Donald Trump's global trade tariffs. Despite five years of dramatic supply disruptions, from the COVID pandemic to today's trade wars, Toro is resisting any temptation to stack its warehouses to the rafters. "We are at probably pre-pandemic inventory levels," says its chief supply-chain manager, Kevin Carpenter, looking relaxed in front of a whiteboard at his office in Minneapolis. "I mean 2019. I think everybody will be at a 2019 level." Among U.S. manufacturers, inventories have roller-coasted this year as they rushed to beat Trump's deadlines for tariff hikes, only to see them repeatedly delayed. But since their post-pandemic expansion, inventories have mostly contracted, according to U.S. Institute for Supply Management data. Instead, "just in time" inventory management - which aims to increase efficiency and reduce waste by ordering goods only as they are needed - is back. But how can firms run lean inventories even as tariffs fluctuate, export bans come out of the blue, and conflict rages? One of the answers, they say, is artificial intelligence. Carpenter says he uses AI to digest the daily stream of news that could impact Toro's business, from Trump's latest social media posts to steel prices, into a custom-made podcast that he listens to each morning. His team also uses generative AI to sieve an ocean of data and to suggest when and how many components to buy from whom. It is a boom industry. Spending on software that includes generative AI for supply chains, capable of learning and even performing tasks on its own, could hit $55 billion by 2029, up from $2.7 billion now, according to U.S. research firm Gartner, driven in part by global uncertainties. HYPE "The tool just puts up in front of you: 'I think you can take 100 tonnes of this product from this plant to transfer it to that plant. And you just hit accept if that makes sense (to you)," McKinsey supply chain consultant Matt Jochim said. The biggest providers of overall supply chain software by revenue are Germany's SAP, U.S. firms Oracle, Coupa and Microsoft and Blue Yonder, a unit of Panasonic, according to Gartner. Generative AI is in its infancy, with most firms still piloting it spending modest amounts, industry experts say. Those investments can climb to tens of millions of dollars when deployed at scale, including the use of tools known as AI agents, which make their own decisions and often need costly upgrades to data management and other IT systems, they said. In commenting for this article, SAP, Oracle, Coupa, Microsoft and Blue Yonder described strong growth for generative AI solutions for supply chains without giving numbers. At U.S. supply chain consultancy GEP, which sells AI tools like this, Trump's tariffs are helping to drive demand. "The tariff volatility has been big," says GEP consultant Mukund Acharya, an expert in retail industry supply chains. SAP said the uncertainty was driving technology take-up. "That's how it was during the financial crisis, Brexit and COVID. And it's what we're seeing now," Richard Howells, SAP vice president and supply chain specialist, said in a statement. An AI agent can sift real-time news feeds on changing tariff scenarios, assess contract renewal dates and a myriad of other data points and come up with a suggested plan of action. But supply chain experts warn of AI hype, saying a lot of money will be wasted on a vain hope that AI can work miracles. "AI is really a powerful enabler for supply chain resilience, but it's not a silver bullet," says Minna Aila, communications chief at Finnish crane-maker Konecranes and member of a business board that advises the OECD on issues including supply chain resilience. "I'm still looking forward to the day when AI can predict terrorist attacks that are at sea, for instance." Konecranes' logistics partners are deploying AI on more mundane data, like weather forecasts. The company makes port cranes that are up to 106 metres (348 ft) high when assembled. When shipping them, AI marries weather forecasts with data like bridge heights to optimise the route. "To ship those across oceans, you do have to take into consideration weather," Aila says. RISING COSTS By keeping inventories low, firms can bolster profit margins that are under pressure from rising costs. Every component or finished product sitting on a shelf is capital tied up, incurring finance and storage costs and at risk of obsolescence. McKinsey has been surveying supply-chain executives since the pandemic. Its most recent survey showed that respondents relying on bigger inventory to cushion disruptions fell to 34 per cent last year from 60 per cent in 2022. Early responses from its upcoming 2025 survey suggest a similar picture, Jochim said. Gartner supply chain analyst Noha Tohamy says that without AI, companies would be slower to react and be more likely to be drawn into building up inventories. "When supply chain organisations don't have that visibility and don't really understand the uncertainty, we go for inventory buffering," Tohamy says. But AI agents won't put supply chain managers out of work, not yet, consultants say. Humans still need to make strategic and big tactical decisions, leaving AI agents to do more routine tasks like ordering and scheduling production maintenance. Toro supply chain chief Carpenter says that without AI, supply chain managers might need to run bigger teams as well.
CNA
3 hours ago
- CNA
CoreWeave shares drop as growing losses eclipse AI demand surge
CoreWeave shares sank 10 per cent in premarket trading on Wednesday after the Nvidia-backed company posted a bigger-than-expected loss, raising doubts about its ability to keep costs under check amid robust AI demand. Its results underscore the tension between its rapid revenue growth and mounting financial strain as operating expenses jumped nearly fourfold to $1.19 billion in the second quarter. Analysts have turned skeptical of CoreWeave's excessive dependence on certain customers and its ability to grow profitably due to widening losses, heavy capital needs and deteriorating debt coverage. The company on Tuesday posted a net loss of $290.5 million, compared with analysts' average estimate of $190.6 million, according to data compiled by LSEG. "CoreWeave does not currently generate enough profit to pay all its debt holders, certainly not equity holders," D.A. Davidson analysts said. The company had around $8 billion in debt as of last year and had said in March that it would use about $1 billion of the IPO proceeds to repay debt. CEO Michael Intrator said the company is scaling rapidly to meet "unprecedented demand for AI", but noted that "accessing power shells capable of delivering the scale of infrastructure our clients require" remains the biggest constraint. "This continues to be a business that is not worth scaling," D.A. Davidson analysts said. CoreWeave operates 33 AI data centers in the U.S. and Europe, offering access to backer Nvidia's GPUs, which are highly sought after for training and running large AI models. Surging demand for its AI infrastructure helped the company top quarterly revenue estimates. Its stock price has jumped nearly three-fold since its IPO in March. Investors will seek clarity on the lock-up period related to the company's IPO, which could expire later this week compared to the typical six-month time frame. Sometimes, a stock is pressured when the lock-up ceases.
Business Times
3 hours ago
- Business Times
US embeds trackers in AI chip shipments to catch diversions to China: sources
[SINGAPORE / NEW YORK] US authorities have secretly placed location tracking devices in targeted shipments of advanced chips they see as being at high risk of illegal diversion to China, according to two people with direct knowledge of the previously unreported law enforcement tactic. The measures aim to detect AI chips being diverted to destinations which are under US export restrictions, and apply only to select shipments under investigation, the people said. They show the lengths to which the US has gone to enforce its chip export restrictions on China, even as the Trump administration has sought to relax some curbs on Chinese access to advanced American semiconductors. The trackers can help build cases against people and companies who profit from violating US export controls, said the people who declined to be named because of the sensitivity of the issue. Location trackers are a decades-old investigative tool used by US law enforcement agencies to track products subject to export restrictions, such as airplane parts. They have been used to combat the illegal diversion of semiconductors in recent years, one source said. Five other people actively involved in the AI server supply chain say they are aware of the use of the trackers in shipments of servers from manufacturers such as Dell and Super Micro, which include chips from Nvidia and AMD . BT in your inbox Start and end each day with the latest news stories and analyses delivered straight to your inbox. Sign Up Sign Up Those people said the trackers are typically hidden in the packaging of the server shipments. They did not know which parties were involved in installing them and where along the shipping route they were put in. Reuters was not able to determine how often the trackers have been used in chip related investigations or when US authorities started using them to investigate chip smuggling. The US started restricting the sale of advanced chips by Nvidia, AMD and other manufacturers to China in 2022. In one 2024 case described by two of the people involved in the server supply chain, a shipment of Dell servers with Nvidia chips included both large trackers on the shipping boxes and smaller, more discreet devices hidden inside the packaging – and even within the servers themselves. A third person said they had seen images and videos of trackers being removed by other chip resellers from Dell and Super Micro servers. The person said some of the larger trackers were roughly the size of a smartphone. The US Department of Commerce's Bureau of Industry and Security, which oversees export controls and enforcement, is typically involved, and Homeland Security Investigations and the Federal Bureau of Investigation, may take part too, said the sources. The HSI and FBI both declined to comment. The Commerce Department did not respond to requests for comment. The Chinese foreign ministry did not have immediate comment. Super Micro said in a statement that it does not disclose its 'security practices and policies in place to protect our worldwide operations, partners, and customers.' It declined to comment on any tracking actions by US authorities. Dell said it is 'not aware of a US government initiative to place trackers in its product shipments.' Nvidia declined to comment, while AMD did not answer a request for comment. Chip restrictions The US, which dominates the global AI chip supply chain, has sought to limit exports of chips and other technology to China in recent years to restrain its military modernisation. It has also put restrictions on the sale of chips to Russia to undercut war efforts against Ukraine. The White House and both houses of Congress have proposed requiring US chip firms to include location verification technology with their chips to prevent them from being diverted to countries where US export regulations restrict sales. China has slammed the US exports curbs as part of a campaign to suppress its rise and criticised the location tracking proposal. Last month, the country's powerful cyberspace regulator summoned Nvidia to a meeting to express its concerns over the risks of its chips containing 'backdoors' that would allow remote access or control, which the company has strongly denied. In January, Reuters reported the US had traced organised AI chip smuggling to China via countries such as Malaysia, Singapore, and the UAE – but it is unclear if tracking devices were involved. The use of trackers by US law enforcement goes back decades. In 1985, Hughes Aircraft shipped equipment subject to US export controls, according to a court decision reviewed by Reuters. Executing a search warrant, the US Customs Service intercepted the crate at a Houston airport and installed a tracking device, the decision noted. US export enforcement agents sometimes install trackers after getting administrative approval. Other times they get a judge to issue a warrant authorising use of the device, one source said. With a warrant, it is easier to use the information as evidence in a criminal case. A company may be told about the tracker, if they are not a subject of the investigation, and may consent to the government's installation of the trackers, the source added. But the devices can also be installed without their knowledge. People involved in diverting export-controlled chip and server shipments to China said they were aware of the devices. Two of the supply chain sources, who are China-based resellers of export-controlled chips, said they regularly took care to inspect diverted shipments of AI chip servers for the trackers due to the risks of the devices being embedded. An affidavit filed with a US Department of Justice complaint regarding the arrests of two Chinese nationals charged with illegally shipping tens of millions of US dollars' worth of AI chips to China earlier this month describes one co-conspirator instructing another to check for trackers on Quanta H200 servers, which contain Nvidia chips. It said the English language text was sent by a co-conspirator, whose name was redacted, to one of the defendants, Yang Shiwei. 'Pay attention to see if there is a tracker on it, you must look for it carefully,' said the person, who went on to call the Trump administration by an obscenity. 'Who knows what they will do.' REUTERS
