Sam-Altman-says-OpenAI-will-tweak-its-Pentagon-deal-after.jpeg

Sam Altman says OpenAI will tweak its Pentagon deal after surveillance backlash

OpenAI said it is amending its contract with the Pentagon.

After public concerns that OpenAI’s new deal with the Pentagon would allow the government to use its AI for mass surveillance, CEO Sam Altman posted an internal memo to X on Monday evening, saying that the company is working with the Pentagon to “make some additions in our agreement.”

“Consistent with applicable laws, including the Fourth Amendment to the United States Constitution, National Security Act of 1947, FISA Act of 1978, the AI system shall not be intentionally used for domestic surveillance of US persons and nationals,” Altman wrote on X.

“The Department also affirmed that our services will not be used by Department of War intelligence agencies (for example, the NSA). Any services to those agencies would require a follow-on modification to our contract,” Altman added.

Altman’s memo came after OpenAI struck a deal with the Pentagon on Friday to deploy its AI models on classified military networks. The contract stepped into a standoff between the Pentagon and Anthropic and happened just a day before the US struck Iran.

In his note, Altman said that he got things “wrong,” saying the company should not have “rushed” to seal the deal.

“The issues are super complex, and demand clear communication,” he said. “We were genuinely trying to de-escalate things and avoid a much worse outcome, but I think it just looked opportunistic and sloppy.”

Hours before the OpenAI deal was announced, President Donald Trump ordered federal agencies to halt use of Anthropic’s Claude system, following a breakdown in talks over the military use of AI. Anthropic had specific red lines: explicit contractual bans on mass domestic surveillance and fully autonomous weapons, which are systems capable of killing without human oversight.

As of Friday, nearly 500 OpenAI and Google employees signed on to an open letter in support of Anthropic’s decision.

The OpenAI deal soon triggered backlash and concerns that OpenAI’s tools would be used for domestic surveillance or for lethal autonomous weapons, claims which Altman immediately disputed. Protests took place in front of the OpenAI office in San Francisco and London, and QuitGPT, an advocacy group against OpenAI, has launched a boycott and organized a protest scheduled for Tuesday.

Anthropic did not immediately respond to a request for comments.




Source link

Katherine Tangalakis-Lippert's face on a white background

OpenAI shares its contract language and ‘red lines’ in agreement with the Department of War

OpenAI says its agreement with the Department of War is “better” and has more safety guardrails than the one Anthropic was blacklisted for refusing to comply with.

In a blog post published Saturday, OpenAI shared some contract language from its agreement with the Department of War, including clauses that indicate its tech cannot be used for mass domestic surveillance or to power autonomous weapons or high-stakes decision systems like “social credit” scores.

“We think our agreement has more guardrails than any previous agreement for classified AI deployments, including Anthropic’s,” OpenAI’s post read. “In our agreement, we protect our red lines through a more expansive, multi-layered approach. We retain full discretion over our safety stack, we deploy via cloud, cleared OpenAI personnel are in the loop, and we have strong contractual protections. This is all in addition to the strong existing protections in U.S. law.”

OpenAI CEO Sam Altman took to social media shortly after the company’s blog post was published, answering questions from users concerned about the nature of OpenAI’s agreement with the government.

In Ask-Me-Anything-style responses, he doubled down on OpenAI’s agreement being better than Anthropic’s, not just for the broader AI landscape but also for the American people.

“Anthropic seemed more focused on specific prohibitions in the contract, rather than citing applicable laws, which we felt comfortable with,” Altman wrote in response to a question about why OpenAI agreed to partner with the government when its rival would not. “I think Anthropic may have wanted more operational control than we did.”

OpenAI’s agreement with the federal government comes on the heels of Anthropic being blacklisted and declared a supply chain risk after refusing to comply with the military’s terms of use for the company’s frontier model, Claude.

Anthropic, in a Friday statement, said that “no amount of intimidation or punishment from the Department of War will change our position on mass domestic surveillance or fully autonomous weapons” and vowed to “challenge any supply chain risk designation in court.”

OpenAI, in its Saturday post, argued that Anthropic should not be designated as a supply chain risk and said it had made its position “clear to the government.” Its agreement with the Department of War stemmed, in part, from a desire to “de-escalate things between DoW and the US AI labs.”

“A good future is going to require real and deep collaboration between the government and the AI labs,” OpenAI’s post reads. “As part of our deal here, we asked that the same terms be made available to all AI labs, and specifically that the government would try to resolve things with Anthropic; the current state is a very bad way to kick off this next phase of collaboration between the government and AI labs.”

Representatives for OpenAI and Anthropic did not immediately respond to requests for comment from Business Insider. It was not immediately clear whether Anthropic, or any other leading AI company, had been offered similar contractual terms to those that OpenAI said it had agreed to.

OpenAI said that, as part of its deal with the Department of War, it will maintain “full control” over the safety stack it deploys, and robust “safety guardrails” to prevent misuse. Should the government violate the terms of the agreement, OpenAI said it “could” terminate the contract.

“We don’t expect that to happen,” OpenAI said in its post.

Altman, in his Ask Me Anything posts, wrote that OpenAI would not agree to allow the government to use its technology for mass domestic surveillance “because it violates the constitution.”

He added that he is prepared for a potential dispute over the legality of specific governmental requests in the future, but added that if the Constitution were amended to make such surveillance legal, “Maybe I would quit my job.”

“I very deeply believe in the democratic process, and that our elected leaders have the power, and that we all have to uphold the constitution,” Altman wrote. “I am terrified of a world where AI companies act like they have more power than the government. I would also be terrified of a world where our government decided mass domestic surveillance was ok. I don’t know how I’d come to work every day if that were the state of the country/Constitution.”

The dispute between the government and the AI giants has sparked widespread criticism, with critics concerned about the ethical implications of the Department of War’s use of AI and OpenAI’s agreement to provide the government access to its technology.

OpenAI on Saturday said it believes AI will “introduce new risks in the world” and, by allowing the government use of its models, will give people defending national security “the best tools” to do so.

Business Insider previously reported that Anthropic’s model, Claude, shot to the top of the app store on Saturday, and many people on social media, including celebrities like Katy Perry, have publicly posted about canceling their ChatGPT subscriptions in the wake of OpenAI’s agreement with the government.




Source link

OpenAI-just-hired-another-employee-from-Mira-Muratis-Thinking-Machines.jpeg

OpenAI just hired another employee from Mira Murati’s Thinking Machines Lab

Another employee at Thinking Machines Lab is leaving to rejoin OpenAI.

It’s the latest in a string of departures from the $12 billion AI startup, which is led by former OpenAI CTO Mira Murati and lately has been the subject of high-profile poaching campaigns from bigger tech companies.

The latest employee to go back to OpenAI is Jolene Parish, who joined Thinking Machines Lab in April last year, according to her LinkedIn profile. She had worked at OpenAI for three years prior. Before that, she worked for 10 years on security at Apple, her profile says.

Other employees rejoined OpenAI last month. Two co-founders, former CTO Barret Zoph and Luke Metz, both left, along with researcher Sam Schoenholz.

Lia Guy, another researcher, also rejoined OpenAI, The Information reported. Another cofounder, Andrew Tulloch, left for Meta late last year, The Wall Street Journal reported.

OpenAI and Thinking Machines Lab declined to comment.

Thinking Machines Lab raised a monster $2 billion funding round last year, valuing the company at $12 billion, spokespeople said at the time. The startup launched its first product, Tinker, last October.

The San Francisco-based company has become known for attracting star-studded talent. It quietly hired Neal Wu, a legendary coder who won three gold medals in an Olympiad for programming, and Soumith Chintala, the creator of the open-source AI project PyTorch at Meta, who is now Thinking Machines Lab’s CTO, Business Insider previously reported.

Have a tip? Contact this reporter via email at crollet@businessinsider.com or on Signal and WhatsApp at 628-282-2811. Use a personal email address, a nonwork WiFi network, and a nonwork device; heres our guide to sharing information securely.




Source link

OpenAI-Meta-and-Apples-latest-battle-Breaking-your-phone-addiction.jpeg

OpenAI, Meta, and Apple’s latest battle: Breaking your phone addiction

The average American picks up their phone more than 200 times a day. Teens are pinged with some 250 notifications a day — during school, after school, and overnight. The apps meant to prevent you from checking apps have done little to stop the problem. Now, some of the tech companies that helped create our screen dependence are trying to disrupt it.

Later this year, OpenAI plans to debut a small, screenless device that Sam Altman describes as more “peaceful” than a smartphone. Apple, the Oz of screentime, is developing smart glasses, a pin, and AirPods with more AI built in, according to a Tuesday report from Bloomberg, with the rumored pendants featuring microphones and cameras to be the “eyes and ears” of the iPhone. Meta has teased its fully augmented reality Orion glasses since 2024. While that device doesn’t have a release date, the company last year sold some 7 million pairs of its smart glasses, which is the start of the post-smartphone future Mark Zuckerberg has predicted. Eventual smart specs could be more screen all-the-time than screenless, but they also rely on AI to make the experience much more hands-free than swiping and scrolling on a phone.

Could AI be what finally breaks our phone addiction?

Since 2007, no device out of Silicon Valley has captured universal imagination the way Steve Jobs did when he put your iPod, your phone, and the internet together on a 3.5-inch screen. Competitors have tried for a decade-plus to get people to shift us from the iPhone to smart glasses, and largely failed. The awe around smartphones has turned to derision, as excessive screen time is linked to disrupted sleep, anxiety, and fractured attention. Now, developers are hoping the AI boom can give us the next big thing.

Beating the smartphone would mean replacing a device that 91% of American adults now carry — a device for which millions of apps have been developed and people now depend on in lieu of wallets and cameras and health monitors. New AI devices can’t just copy what smartphones do, says Ramon Llamas, a research director at a technology intelligence firm IDC: They have to show they have a solution to an everyday problem. If they don’t, Llama says, “these things are just gonna really end up as solutions looking for a problem to solve.”


Critiques of screen time can be as blunt and smoothbrained as what the critics say excessive screen time makes you. A seven-hour daily log may seem like a staggering amount of dependence, but what did the person spend those seven hours doing? Doomscrolling late into the night, or FaceTiming with a far-away friend? With AI wearables, there’s the risk of becoming dependent on the device for different reasons.

“The screen may not be there, but what’s getting filled in the back is already this problem of AI companionship,” says Olivia Gambelin, an AI ethicist and author of the book “Responsible AI.” An AI device designed to do something very specific — like listen to a meeting and then send follow-up emails or messages related to action points discussed — could save people time and keep them from writing tedious emails and Slack messages from their desk. But that same device listening in to personal conversations with family and friends could compromise a relationship, and erode the positive effects that texting a friend to check-in can have on both people (already, my friends are tiring of AI summaries on the iPhone that summarize our group text and become an intermediary into our threads of gossip and jokes in the name of efficiency). Wearing microphones and cameras to social interactions and into businesses is likely to really weird out some of the people around you. More people are entering into romantic, dependent relationships with AI companions, and a swell of loud dissenters are criticizing the technology for taking jobs and attempting to replicate human relationships.

But OpenAI is betting that it can package its technology in a device in a way that calms the user. “When I use current devices or most applications, I feel like I am walking through Times Square in New York and constantly just dealing with all the little indignities along the way,” Altman said in November. OpenAI’s device, he said, would be less Time Square, more “sitting in the most beautiful cabin by a lake and in the mountains and sort of just enjoying the peace and calm.” That’s because the AI device would learn “contextual awareness of your whole life,” and when best to send you alerts.

The screen itself may not be the problem; it’s what’s summoning us to the screen.

Other AI wearables have failed by falling short of that goal. Humane AI sold a wearable pin, priced at $700 plus a monthly fee to connect it, but pulled it from the market a year ago. It failed perhaps because it tried too hard to replace our phones — it didn’t interact with them, but provided a shoddy replacement. Novelty wasn’t a factor that could outshine usability. The AI Friend pendant, which can’t search the internet or help with tasks outside of sending reminders and acts instead as an eavesdropping sycophant around its user’s neck, was mocked relentlessly and sold just a few thousand devices after it hit the market last year.

Companies trying to make AI hardware should focus on “transformative features,” Jason Low, research director at Omdia, tells me in an email. AI wearables must be more than “marginally more convenient,” should integrate with our existing products, and have a clear, stated value. For example, glasses that provide real-time language translation or devices for fitness and health tracking offer features our smartphones can’t do as well. The Oura ring continues to grow in popularity, particularly among women after starting out as a niche tech bro buy, for the novel insights it can offer; the company announced last fall it has sold 5.5 million rings since 2015, with more than 2.5 million sold between June 2024 and September 2025. “These devices often deliver a more polished user experience compared to general-purpose, do-it-all AI devices,” Low says.

Llamas tells me that the AI functions of a wearable have to be “contextual, personalized, and actionable,” like reminding the wearer to send birthday flowers or responding accurately to being asked to direct the user to the nearest Starbucks. A first attempt device shouldn’t try to replace the smartphone, but to integrate with the Apple or Google ecosystems, he says. Apple and OpenAI did not respond to requests for comment about their rumored products for this story.

If anything has hyped Silicon Valley like the iPhone, it’s been AI. But three years after the mainstream adoption of ChatGPT, the value generative AI in the white collar workforce has yet to be fully realized. That could make a product for consumers a hard sell, too. “Some of the overwhelm that’s coming with AI that I see in general users is you can use it for everything, or it’s promoted that way, which is actually quite stifling,” Gambelin says.

In our quest to find a peaceful equilibrium with tech, the screen itself may not be the problem; it’s what’s summoning us to the screen. Its bright colors, games, and infinite scroll give quick dopamine hits that entice us to stay glued to it. But much of what pings my phone throughout the day are useless notifications trying to get me to reopen one of the dozens of apps — a markdown moment on a clothing thrifting app, a like on the Instagram story I’ve posted of my dog from my best friend, and ironically, a report of how much time I’ve already logged. There’s a relentless business model at play to keep us on these apps. No screens would mean no infinite scroll through TikTok, no Candy Crush — but app developers and companies may need to find new ways to reach people if wearables caught on, and an always-there AI device and companion might not be as peaceful as Altman describes. Our collective screen time is a problem, but the AI wearable will have to surprise us all with something novel to be useful.


Amanda Hoover is a senior correspondent at Business Insider covering the tech industry. She writes about the biggest tech companies and trends.

Business Insider’s Discourse stories provide perspectives on the day’s most pressing issues, informed by analysis, reporting, and expertise.




Source link

OpenClaw-creator-says-Europes-stifling-regulations-are-why-hes-moving.jpeg

OpenClaw creator says Europe’s stifling regulations are why he’s moving to the US to join OpenAI

In Europe, there’s been a lot of handwringing over why there are very few large, successful tech companies in the region. Peter Steinberger, the creator of the agentic AI hit OpenClaw, has an answer.

Steinberger was recently hired by OpenAI and is moving from Europe to the US. An Austrian by birth, he previously split his time between London and Vienna.

On X, a professor from a European university asked why Europe couldn’t retain this tech talent.

Steinberger replied that most people in the US are enthusiastic, while in Europe, he’s scolded about responsibility and regulations.

If he built a company in Europe, he would struggle with strict labor regulations and similar rules, he added.

At OpenAI, he said most employees work 6 to 7 days a week and are paid accordingly. In Europe, that would be illegal, he added.

The most valuable company in Europe is Dutch chip-equipment maker ASML, valued at about $550 billion. In contrast, there are 10 US companies worth more than $1 trillion. Most of these are tech companies.

In 2024, a landmark EU report found that the region had fallen behind the US, particularly in innovation. It proposed a series of changes to tackle the problem, but by the end of 2025, few of the recommendations had been implemented.

Steinberger said he was hopeful about EU INC, an effort to create a single corporate legal framework to make it simpler to run a business across the region.

But this seems to be “fizzling out,” he wrote on X. “Watered down, too much egoistic national interest that ultimately hurts everyone.”

Sign up for BI’s Tech Memo newsletter here. Reach out to me via email at abarr@businessinsider.com.




Source link

Headshot of Ben Shimkus

Elon Musk and OpenAI posture over pizza as the AI talent war heats up

The rivalry between xAI and OpenAI is heating up again — this time, over wood-fired pizza.

Over the weekend, Elon Musk and an OpenAI engineer jockeyed on X about wood-fired crusts, dough fermentation, and campus chefs.

On its face, it was a lighthearted back-and-forth about free pizza for lunch. Underneath, it encapsulates a trend playing out in Silicon Valley: rival AI companies are publicly pitching culture — and perks like free lunch — in the talent war for top engineers.

The exchange began when Musk reposted a video of an xAI engineer calling his job the “opportunity of a lifetime.”

“Join @xAI,” Musk wrote.

The post quickly drew a response from xAI’s competitor, OpenAI.

“Or join Codex,” said Thibault Sottiaux, an engineering lead working on OpenAI’s Codex software agent, who is also hiring. OpenAI operates “with much of the same principles,” he wrote — before adding an increasingly common recruitment pitch.

“Join the bright side, we have pizza,” Sottiaux wrote.

Musk fired back: “But how good is your wood oven pizza?”

The pizza posturing then shifted to ingredients — and the corporate chefs preparing them.

“But how about the dough?” he wrote back. “Can’t take shortcuts, needs 24 hours at least. And our chef is 🔥.”

“Our chef is so good that God looked down at the food from heaven and said you my most delicious creation,” Musk replied.

“And after having a bite, he wasn’t 100% satisfied and asked our chef to improve upon the SoTA,” Sottiaux said. “Our chef delivered, and created a recipe now universally credited to accelerating the AGI timeline.”

The very real fight behind the pizza posts

The tomato pie-based banter was sweet — but the subtext was spicier.

AI labs are locked in a high-stakes dash for elite engineers, with high-end compensation packages stretching into the nine-figure territory.

Companies including Amazon, Microsoft, Meta, OpenAI, and Musk’s xAI are competing for a relatively small pool of researchers capable of building the next generation of models and infrastructure.

Aside from money, two key perks have emerged in the AI talent wars, according to professional AI poacher Mark Zuckerberg: access to GPUs and fewer direct reports.

“People say, ‘I want the fewest number of people reporting to me and the most GPUs,'” Zuckerberg said in 2025 TITV interview.

At the same time, the broader tech industry has pulled back on many of the pre-pandemic perks amid cost-cutting. Remote work has narrowed, layoffs have gathered steam, and perks like pet care stipends and expansive wellness benefits are becoming less common for new hires.

But there’s one perk that has remained: the fancy lunch spread.

Might as well throw in wood-fired pizza, too.




Source link

Sam-Altman-says-OpenClaw-creator-Peter-Steinberger-is-joining-OpenAI.jpeg

Sam Altman says OpenClaw creator Peter Steinberger is joining OpenAI to build next-gen personal agents

  • Sam Altman says OpenClaw creator Peter Steinberger is joining OpenAI.
  • OpenClaw is a viral AI agent launched last month.
  • Altman said Steinberger will build “next generation” AI agents at OpenAI.

OpenAI just scored a win in the AI talent wars.

Sam Altman said Sunday on X that Peter Steinberger, the creator of OpenClaw, the viral AI agent powering the agent-only social network Moltbook, is joining OpenAI.

Altman said Steinberger would build the “next generation” of personal AI agents at the company.

“He is a genius with a lot of amazing ideas about the future of very smart agents interacting with each other to do very useful things for people,” Altman said about Steinberger. “We expect this will quickly become core to our product offerings.”

Altman added that OpenClaw, which was for a brief moment in time known as Moltbot and then Clawdbot before Anthropic took notice, will live on as an open-source project supported by OpenAI.

“The future is going to be extremely multi-agent and it’s important to us to support open source as part of that,” he wrote.

Steinberger, previously best known for founding the PDF processing company PSPDFKit, came out of retirement to launch OpenClaw in late 2025.

He is likely to bring a new perspective to OpenAI’s race to develop artificial general intelligence. Steinberger said he believes AGI is best as a specialized form of intelligence rather than a generalized one.

“What can one human being actually achieve? Do you think one human being could make an iPhone or one human being could go to space?” Steinberger said on a Y Combinator podcast in February. “As a group we specialize, as a larger society we specialize even more.”




Source link

Katherine Li, West Coast breaking news reporter at the Business Insider.

Anthropic and OpenAI release dueling AI models on the same day in an escalating rivalry

The rivalry between OpenAI and Anthropic intensified this week.

The two companies released dueling new AI models on Thursday and had back-to-back podcast appearances on “TBPN.”

On Thursday, Anthropic unveiled Claude Opus 4.6, an upgraded model that the company says would improve performance on office productivity and coding tasks, with an expanded “context window” that allows it to work through longer documents and more complex projects in a single session.

Meanwhile, OpenAI punched back with its own new coding-focused model called GPT-5.3-Codex, which the company says runs faster, uses fewer computing resources, and can generate and manage complex software from English instructions. The new version also comes alongside a stand-alone Codex desktop app.

Both Sam Altman, the OpenAI CEO, and Sholto Douglas, one of Anthropic’s leading researchers, appeared on the “TBPN” podcast in back-to-back chats with show host John Coogan and Jordi Hays.

“I think we will be heading towards a workflow where a lot of people just feel like they’re managing a team of agents,” said Altman. “And as the agents get better, they’ll keep operating at a higher and higher level of abstraction.”

Douglas, who appeared in the subsequent timeslot, told Coogan and Hays that users have been comparing previous Anthropic and OpenAI models, and they have noticed some key differences.

“The OpenAI models were a bit better at trying really, really, really hard on tough problems, but the Anthropic models were much faster and so forth,” Douglas said.

“And so they worked on speed while we worked on making the models much, much better at really, really tough problems,” Douglas added of the Opus 4.6.

The latest release is part of a long-running competition between Anthropic and OpenAI, dating back to 2021, when a group of OpenAI researchers left to form Anthropic, aiming to develop safer and more controlled AI systems.

A big week for Anthropic

This week, Anthropic’s launch of industry-specific plugins triggered a stock market sell-off as Wall Street worried about AI’s impact on software.

Anthropic also took a subtle shot at OpenAI with a series of ads released this week, including one that will air during the Super Bowl.

The ads feature unnamed humanized AIs dropping ads in the middle of their advice, alongside the promise that its model, Claude, will remain ad-free.

OpenAI announced in January that ads are coming to ChatGPT for users of the free version.

Altman subsequently hit back, calling Anthropic “dishonest” and defending ChatGPT as a product that brings AI “to billions of people who can’t pay for subscriptions.” He also clarified that the ads will be “clearly labeled” to differentiate themselves from the chatbot’s answers to queries.

“We are not stupid. We respect our users. We understand that if we did something like what those ads depict, people would rightfully stop using the product,” Altman told the “TBPN” podcast on Thursday.

“Our first principle with ads is that we’re not going to put stuff into the LLM stream,” Altman added. “That would feel crazy dystopic, like a bad sci-fi movie.”




Source link

Chong Ming Lee, Junior News Reporter at Business Insider's Singapore bureau.

Jensen Huang says Nvidia would love to back an OpenAI IPO, and there’s ‘no drama’ with Sam Altman

Jensen Huang says Nvidia would love to invest in a future OpenAI IPO.

Huang said in an interview on CNBC’s “Mad Money” on Tuesday that there was “no drama” between Nvidia and OpenAI CEO Sam Altman, pushing back against recent chatter of tension in the relationship between the two companies.

“The first deal is on,” the Nvidia CEO said, referring to the company’s September deal with OpenAI, under which the company said it planned to invest up to $100 billion in the AI startup.

“​​And then there’s, of course, an IPO in the future,” he added. “We love to be participating in that as well,” he added.

Huang also described OpenAI as a “once in a generation company” and said Nvidia is “delighted to invest in it.”

His comments come amid reports suggesting internal unease around the deal.

The Wall Street Journal reported on Saturday that the investment had sparked internal concerns at Nvidia, with some executives questioning the deal, according to people familiar with the matter.

Separately, Reuters reported on Tuesday that OpenAI had been unhappy with certain newer Nvidia chips and had looked at alternatives since last year, citing people familiar with the matter.

Huang told reporters in Taipei on Saturday that speculation of any dissatisfaction with OpenAI was “nonsense.”

“We will invest a great deal of money, probably the largest investment we’ve ever made,” he added.

Altman has also pushed back on rumors of tension.

“We love working with NVIDIA and they make the best AI chips in the world,” wrote Altman in a post on X on Tuesday.

“We hope to be a gigantic customer for a very long time. I don’t get where all this insanity is coming from,” he added.

OpenAI is one of the world’s most valuable private AI companies and a major customer for Nvidia’s chips, which power the training and deployment of large language models.

The startup has not announced plans for an IPO, but its fundraising and computing needs have fueled speculation about how it will finance future growth.

“Big Short” investor Michael Burry said in a Substack exchange in January that he was surprised that ChatGPT “kicked off a multi-trillion-dollar infrastructure race.”

“It’s like someone built a prototype robot and every business in the world started investing for a robot future,” he wrote.




Source link

Microsoft-says-OpenAI-is-driving-45-of-the-backlog-for.jpeg

Microsoft says OpenAI is driving 45% of the backlog for Azure cloud computing

Microsoft is facing capacity constraints, and OpenAI is driving a large portion of the backlog in its cloud computing business.

The company said its backlog in commercial bookings, a metric referred to as remaining performance obligations, ballooned 110% year over year to $625 billion when it reported earnings for the second quarter on Wednesday.

OpenAI accounts for roughly 45% of those commitments, Microsoft revealed. The company did not say how much OpenAI contributed during the previous quarter.

Some Wall Street analysts on the call expressed concerns about Microsoft’s dependency on OpenAI.

CEO Satya Nadella said acquiring more Azure clients is important to the tech giant, but it can’t come at the expense of neglecting its other services.

“If you think about it, acquiring an Azure customer is super important to us, but so is acquiring an M365 or a GitHub or a Dragon Copilot, which are all, by the way, incremental businesses and TAMs for us,” Nadella said during Microsoft’s second-quarter earnings call. “And so we don’t want to maximize just one business of ours.”

Shares of Microsoft fell more than 6% in after-market trading on Wednesday, even as the tech giant posted an overall earnings beat.

Morgan Stanley’s Keith Weiss said during the call that some on Wall Street may be spooked by slower growth in overall Azure revenue and the increase in capex spending. Microsoft’s capital expenditures rose 66% year over year to $37.5 billion in the second quarter, another record for the company and testament to the sheer amount of money tech companies are spending amid the AI race.

CFO Amy Hood said that Microsoft has to look at many different areas when it allocates the GPUs and CPUs that come online as a result of its capex spending, including investing in the growth of first-party apps like Microsoft Copilot, devoting GPUs to research and development, and the talent they’ve acquired.

“You end up with the remainder going towards serving the Azure capacity that continues to grow in terms of demand,” she said.

Microsoft is not alone in facing capacity issues.

Executives at OpenAI, which has pledged to spend $250 billion on Azure services, have repeatedly said the startup is held back by a lack of compute, forcing tough trade-offs between product and research.

Wednesday’s earnings mark the first quarter since OpenAI completed its restructuring, which included a new agreement with Microsoft, the startup’s largest investor. Microsoft owns 27% of the public benefit corporation.

“It’s a great partnership,” Hood said of Microsoft’s relationship with OpenAI. It’s allowed us to remain a leader in terms of what we’re building and being on the cutting edge of app innovation.”




Source link