विज्ञान

By Noel Randewich (Reuters) -Nvidia was on the verge of becoming the first company worth $5 trillion on Tuesday after the heavyweight chipmaker said it had $500 billion in bookings for its artificial intelligence processors and that it will build seven new supercomputers for the U.S. Department of Energy. Nvidia's stock closed almost 5% higher, adding over $230 billion in market value, bringing its total value to $4.89 trillion after briefly touching $4.94 trillion. CEO Jensen Huang kicked off a keynote address at a developer conference in the U.S. capital on Tuesday by praising policy by U.S. President Donald Trump while announcing new products and deals. Nvidia is at the core of the global rollout of AI, and is striking deals while navigating a U.S.-China trade war that could determine which technology from the two countries is most used around the world. The Santa Clara, California, company's stock has surged 50% in 2025. Its market capitalization closed above $4 trillion for the first time in July. Microsoft, the world's second most valuable company, rose 2%, lifting its market capitalization to $4.03 trillion. It and OpenAI announced a restructuring deal that frees the ChatGPT maker to move away from its nonprofit roots and likely go public. The supercomputers Nvidia is building for the Energy Department will in part help the U.S. maintain and develop its nuclear weapons arsenal. The largest of the supercomputers will be built with Oracle and contain 100,000 of Nvidia's high-end Blackwell AI chips. (Reporting by Noel Randewich; Editing by Richard Chang)

(The article has been published through a syndicated feed. Except for the headline, the content has been published verbatim. Liability lies with original publisher.)

(Reuters) -Seagate Technology forecast second-quarter revenue and profit above analysts' estimates on Tuesday, betting on robust demand for its storage devices as cloud providers expand investments in hardware to support generative AI development.  Shares of the company rose over 3% in extended trading.  Cloud providers are pouring hundreds of billions of dollars into hardware that aids in the data-intensive process of training and running GenAI applications. Seagate, which supplies disk drives used to store vast amounts of data, has benefited from this.  Seagate expects second-quarter revenue of $2.70 billion, plus or minus $100 million, compared with analysts' average estimate of $2.66 billion, according to data compiled by LSEG. It expects adjusted profit of about $2.75 per share in the current quarter, above estimates of profit of $2.62 apiece. Boosted shipments of its products are being driven by "demand for existing use cases such as social media video platforms as well as growth driven by new AI applications," CEO Dave Mosley said in a statement. The optimistic forecast represents further normalization in memory device inventory levels, after a supply glut over the last two years drove cloud providers to cut back on orders in a bid to clear existing inventory. The addition of AI features to personal electronics has boosted the memory requirements of such devices. This will help Seagate, whose hard disk drives and other storage devices are employed in personal electronics.  Seagate reported revenue of $2.63 billion for the first quarter ended October 3, compared with the consensus estimate of $2.55 billion. The company posted adjusted earnings of $2.61 per share versus analysts' expectations of profit of $2.37 apiece. (Reporting by Arsheeya Bajwa; Editing by Alan Barona)

(The article has been published through a syndicated feed. Except for the headline, the content has been published verbatim. Liability lies with original publisher.)

By John Revill ZURICH (Reuters) -Logitech on Tuesday reported better-than-expected sales and earnings for the second quarter as the computer peripherals maker sold more of its artificial intelligence-enabled products. The Swiss-American company said its sales rose 6% to $1.19 billion for the three months ended September 30. The figure came in slightly ahead of the analyst estimate of $1.18 billion, according to Visible Alpha. Meanwhile, non-GAAP operating income rose 19% to $230 million, beating analysts' estimates of $196 million. Logitech said it was being helped by increased sales to businesses, which were investing more in products such as video conferencing equipment and computer peripherals to help with hybrid work. Sales of computer keyboards rose 12%, while those of mice jumped 13% in the quarter. Sales of video collaboration devices rose 5%. Gaming product sales rose 8%, helped by 12 new launches, including steering wheels and special mice aimed at computer gamers. "We delivered another strong quarter, driving growth and excellent profitability," CEO Hanneke Faber said in a statement. The company saw strong demand from businesses and consumers, Faber added. Logitech – which is based in Lausanne, Switzerland, and San Jose, California – has recently been shifting production lines from China to lessen the impact of U.S. tariffs, while its plant in Suzhou will continue to supply China. Logitech also increased prices by 10% in the U.S. in April to counter the tariff impact. For the coming quarter, Logitech expects sales of $1.38 billion to $1.42 billion, an increase of 3%-6% when converted to U.S. dollars. It also said it expected non-GAAP operating income of $270 million to $290 million. (Reporting by John Revill; Editing by Anil D'Silva)

(The article has been published through a syndicated feed. Except for the headline, the content has been published verbatim. Liability lies with original publisher.)

(Removes reference to vehicle launching next year in paragraph 1) (Reuters) -Lucid Group said on Tuesday it plans to launch a mid-size Level 4 self-driving electric vehicle using computer systems from Nvidia, marking a step toward cars that can operate without driver input in most conditions. Automakers are developing AI-based self-driving technology with a focus on improving safety and reducing the need for human control or monitoring in most conditions. Lucid's upcoming car will require minimal human intervention, making it one of the closest to fully automated Level 5 technology. The mid-size model is expected to sit below the Lucid Air sedan in price and size, targeting a broader market segment. The launch is part of Lucid's plan to expand beyond luxury sedans into higher-volume electric cars. Lucid plans to use a combination of sensors, including cameras, radar and lidar, to enable its self-driving system. Market leader Tesla relies solely on camera-based vision technology for its driver-assistance features. The new model is expected to compete with Tesla's best-selling Model Y and Rivian's R2 SUV. Lucid, backed by Saudi Arabia's Public Investment Fund, said its next generation of cars will use Nvidia's DRIVE AGX Thor chips to power their driving systems. The EV maker also plans to use Nvidia's software tools in its factories to improve quality and speed up production. Commercializing self-driving tech has been much harder than anticipated, with high costs, tight regulations and federal investigations forcing many, including General Motors' Cruise, to shut down. Lucid's announcement builds on its efforts to expand its technology partnerships. Earlier this year, it began working with Uber and self-driving technology firm Nuro to provide premium electric cars for ride-hailing.  (Reporting by Akash Sriram in Bengaluru; Editing by Leroy Leo)

(The article has been published through a syndicated feed. Except for the headline, the content has been published verbatim. Liability lies with original publisher.)

(Reuters) -Trump Media and Technology Group said on Tuesday it will introduce prediction markets on its social media platform Truth Social through a partnership with Crypto.com, as event-driven markets continue to make inroads into mainstream finance. The collaboration, involving streaming platform Truth+, fintech brand Truth.Fi and exchange and clearinghouse Crypto.com, will begin beta testing live on Truth Social in the near future, followed by a full launch in the U.S. Trump Media plans to launch the service globally once all the requisite requirements are met. Prediction markets allow users to profit from forecasting a wide range of events, including sports, entertainment, politics, and economic trends. Trading in these event-based contracts has gained momentum since last year's U.S. presidential election. And as interest grows, prediction markets are increasingly seen as a serious contender for broader acceptance in financial circles, with some experts touting them as more accurate than traditional polls, particularly in tightly contested scenarios. Critics, however, argue that these markets are little more than "digital casinos". "For too long, global elites have closely controlled these markets — with Truth Predict, we're democratizing information and empowering everyday Americans to harness the wisdom of the crowd, turning free speech into actionable foresight," said Devin Nunes, Chairman and CEO of Trump Media, said. Earlier this year, Trump Media and Crypto.com had agreed to a deal with a blank-check acquisition company to launch a new venture that will pursue a treasury-style strategy to accumulate the cryptocurrency platform's native token CRO, deepening the ties of U.S. President Donald Trump to the industry. "Prediction markets are poised to be a multi-deca-billion dollar industry," said Crypto.com CEO Kris Marszalek. Kalshi and Polymarket, two of the largest prediction markets globally, have received a lot of interest from venture capital firms and other investors in recent months, solidifying the reach and engagement of prediction markets. Polymarket had secured an investment from Trump Jr-backed 1789 Capital earlier in the year. (Reporting by Pritam Biswas in Bengaluru; Editing by Shilpi Majumdar)

(The article has been published through a syndicated feed. Except for the headline, the content has been published verbatim. Liability lies with original publisher.)

By Deepa Seetharaman, Stephen Nellis and Deborah Mary Sophia (Reuters) -Microsoft and OpenAI announced a restructuring deal on Tuesday that frees the ChatGPT maker to move away from its nonprofit roots and likely go public so it can finance CEO Sam Altman's ambitious plans to develop data centers and cutting-edge technology.  The deal refashions the $500 billion company as a public benefit corporation that is controlled by a nonprofit with a stake in OpenAI's financial success. In a livestream broadcast, Altman said an initial public offering was the most likely path for the company's future, given how much money is needed to train and build AI systems like the one that supports ChatGPT.  Altman and chief scientist Jakub Pachocki opened the livestream by outlining plans to shift OpenAI from a product company into a platform that lets others build their own tools, services, and businesses on top of its technology. "We can now take this technology and this user base and this sort of framework we've built and get the whole world to build amazing new companies and services and applications on top of it," Altman said. CAPITAL CONSTRAINTS CAUSED TENSIONS Tuesday's agreement removes major constraints on OpenAI's ability to raise capital and secure computing resources from Microsoft that have been in place since the two companies forged their agreement in 2019. Once AI assistant ChatGPT exploded in popularity three years ago, those limitations sparked tensions between the firms. The discussions to restructure OpenAI started after Altman was pushed out of OpenAI temporarily. The episode in late 2023 highlighted how the startup's unusual structure limited the power of its investors and business partners, including Microsoft.  Altman will not get equity in the restructured company, an OpenAI spokesperson said, in a reversal from discussions last year that would have given him a stake. The spokesperson added there will be no changes to his compensation from OpenAI, which pays him some $76,000 a year.  Altman said OpenAI has financial obligations of $1.4 trillion to help build out roughly 30 gigawatts of data-center infrastructure over the next few years. But Altman made clear this was just the beginning, especially if some of OpenAI's more ambitious research goals bear fruit.  Eventually, OpenAI would like to build a data center that can produce one new gigawatt of power per week, Altman said, without offering more details. Each gigawatt costs as much as $50 billion to build, and Altman said he would like to help bring the cost down to $20 billion.   MICROSOFT TO KEEP 27% STAKE The new OpenAI Group PBC will function more like a traditional company, cementing Altman's power and giving him more leeway to strike deals, raise money and shape the broader AI industry. Microsoft will not have rights to any consumer hardware produced by OpenAI, and no longer has the right of first refusal to be OpenAI's compute provider.  Even so, Microsoft will hold a 27% stake in OpenAI and remain a significant player in its future. OpenAI will continue to share about 20% of its revenue with Microsoft for years to come, people familiar with the matter said.   Microsoft's previous 2019 agreement had many provisions that rested on when OpenAI reached that point, and the new deal requires an independent panel to verify OpenAI's claims that it has reached AGI. Microsoft will still hold a stake of about $135 billion in OpenAI Group PBC, which will be controlled by the nonprofit OpenAI Foundation, the companies said.  The Redmond, Washington-based firm has invested $13.8 billion in OpenAI, with Tuesday's deal implying that Microsoft had generated a return of nearly 10 times its investment. Microsoft shares rose 2.4%, sending its market value above $4 trillion again.   The deal keeps the two firms intertwined until at least 2032, with a massive cloud computing contract and with Microsoft retaining some rights to OpenAI products and AI models until then, even if OpenAI reaches artificial general intelligence (AGI), the point at which AI systems can match a well-educated human adult. OpenAI and Microsoft expect to end the revenue-share agreement once an independent panel declares AGI, but OpenAI can also make those payments later. Both companies declined to say when they expect those commitments to end.  With some 800 million weekly users as of October, ChatGPT has exploded in popularity to become the face of AI for many consumers after OpenAI's founding as a nonprofit AI safety group. As the company grew, the Microsoft deal constrained OpenAI's ability to raise funds from outside investors and secure computing contracts as the crush of ChatGPT users and its research into new models caused its computing needs to skyrocket. The requirement that OpenAI exclusively rely on Microsoft for computational resources was among the largest sources of tension.  "OpenAI has completed its recapitalization, simplifying its corporate structure," Bret Taylor, the OpenAI Foundation's board chair, said in a blog post. "The nonprofit remains in control of the for-profit, and now has a direct path to major resources before AGI arrives." Taylor and other board members – including Altman – have the power to appoint and remove members of the PBC's board. The OpenAI Foundation has a 26% stake in OpenAI Group and a warrant to receive additional shares if the company hits certain milestones.  "OpenAI still faces ongoing scrutiny around transparency, data usage, and safety oversight. But overall, this structure should provide a clearer path forward for innovation and accountability," said Adam Sarhan, CEO of 50 Park Investments. Microsoft also said that it has secured a deal with OpenAI where the ChatGPT maker will purchase $250 billion of Azure cloud computing services. In exchange, Microsoft will no longer have the right of first refusal to provide computing services to OpenAI. (Reporting by Deborah Sophia in Bengaluru and Stephen Nellis in Washington; additional reporting by Krystal Hu; Editing by Devika Syamnath, Franklin Paul, Rod Nickel)

(The article has been published through a syndicated feed. Except for the headline, the content has been published verbatim. Liability lies with original publisher.)

(Reuters) -Stellantis said on Tuesday it has partnered with Nvidia, Uber and Foxconn to explore the joint development of autonomous vehicles, aiming to tap into the booming demand for self-driving cars. The collaboration seeks to integrate Nvidia's autonomous driving software and AI computing, Uber's ride-hailing operations and Foxconn's electronics with Stellantis' manufacturing capability to deploy robotaxis with level 4 autonomy. The companies aim to start production in 2028, with Uber deploying 5,000 of Stellantis' autonomous vehicles in select cities worldwide after initial operations begin in the U.S. Earlier in the day, Nvidia unveiled a new self-driving car technology platform called Hyperion, while CEO Jensen Huang announced a partnership with Uber to create a network of robotaxis. (Reporting by Utkarsh Shetti in Bengaluru; Editing by Shilpi Majumdar)

(The article has been published through a syndicated feed. Except for the headline, the content has been published verbatim. Liability lies with original publisher.)

By David Shepardson WASHINGTON (Reuters) -The United States tightened rules on telecoms gear made by Chinese companies deemed a national security risk, the latest move in a broader crackdown on Beijing. The Federal Communications Commission voted 3-0 on Tuesday to block new approvals for devices with parts from companies on its "Covered List" and to allow the agency to bar previously approved equipment in certain cases. "These present loopholes that bad actors could use to threaten the security of our networks," FCC Chair Brendan Carr said. "America's foreign adversaries are constantly looking for ways to exploit any vulnerabilities in our system." The telecoms regulator previously named companies including Huawei, ZTE, China Mobile and China Telecom to the so-called "Covered List," which bars the FCC from authorizing the import or sale of new equipment from those companies. This month Carr said major U.S. online retail websites had removed several million listings for prohibited Chinese electronics as part of a crackdown by the agency. The items removed were on the list or were not authorized by the agency, such as home security cameras and smart watches from companies including Huawei, Hangzhou Hikvision, ZTE and Dahua Technology Co. Hikvision opposes extending restrictions to previously approved models, arguing the FCC is overstepping its authority. "This order is not based on any product-specific evidence," Hikvision said, adding it will "impose unnecessary harm on U.S. communities and small businesses that rely on safe, compliant, and already authorized security systems." The FCC issued a national security notice this month reminding companies of prohibited items including video surveillance equipment. The FCC on October 15 said it was moving to revoke the authority of leading Hong Kong telecom carrier HKT, a subsidiary of PCCW, to operate in the United States, citing national security concerns. In March, the FCC said it was investigating nine Chinese companies on the Covered List including Huawei, ZTE,  Hytera Communications, Dahua Technology Co, Pacific Networks Corp and its ComNet subsidiary and China Unicom (Americas). The FCC previously barred some Chinese companies from providing telecommunications services in the United States, citing national security concerns. Last month, the FCC began proceedings to withdraw recognition from seven test labs owned or controlled by the Chinese government, citing national security concerns. (Reporting by David Shepardson; Editing by Andrea Ricci and Howard Goller)

(The article has been published through a syndicated feed. Except for the headline, the content has been published verbatim. Liability lies with original publisher.)

By David Shepardson WASHINGTON (Reuters) -The United States tightened rules on telecoms gear made by Chinese companies deemed a national security risk, the latest move in a broader crackdown on Beijing. The Federal Communications Commission voted 3-0 on Tuesday to block new approvals for devices with parts from companies on its "Covered List" and to allow the agency to bar previously approved equipment in certain cases. "These present loopholes that bad actors could use to threaten the security of our networks," FCC Chair Brendan Carr said. "America's foreign adversaries are constantly looking for ways to exploit any vulnerabilities in our system." The telecoms regulator previously named companies including Huawei, ZTE, China Mobile and China Telecom to the so-called "Covered List," which bars the FCC from authorizing the import or sale of new equipment from those companies. This month Carr said major U.S. online retail websites had removed several million listings for prohibited Chinese electronics as part of a crackdown by the agency. The items removed were on the list or were not authorized by the agency, such as home security cameras and smart watches from companies including Huawei, Hangzhou Hikvision, ZTE and Dahua Technology Co. Hikvision opposes extending restrictions to previously approved models, arguing the FCC is overstepping its authority. "This order is not based on any product-specific evidence," Hikvision said, adding it will "impose unnecessary harm on U.S. communities and small businesses that rely on safe, compliant, and already authorized security systems." The FCC issued a national security notice this month reminding companies of prohibited items including video surveillance equipment. The FCC on October 15 said it was moving to revoke the authority of leading Hong Kong telecom carrier HKT, a subsidiary of PCCW, to operate in the United States, citing national security concerns. In March, the FCC said it was investigating nine Chinese companies on the Covered List including Huawei, ZTE,  Hytera Communications, Dahua Technology Co, Pacific Networks Corp and its ComNet subsidiary and China Unicom (Americas). The FCC previously barred some Chinese companies from providing telecommunications services in the United States, citing national security concerns. Last month, the FCC began proceedings to withdraw recognition from seven test labs owned or controlled by the Chinese government, citing national security concerns. (Reporting by David Shepardson; Editing by Andrea Ricci and Howard Goller)

(The article has been published through a syndicated feed. Except for the headline, the content has been published verbatim. Liability lies with original publisher.)

By Supantha Mukherjee and Louise Rasmussen STOCKHOLM (Reuters) -Nvidia will pay $1 billion for a 2.9% Nokia stake, the Finnish telecom equipment maker said on Tuesday, propelling its shares to their highest level in nearly a decade. Nokia said the companies will collaborate on artificial intelligence networking solutions and explore opportunities to include its data center communications products in Nvidia's future AI infrastructure plans. Capital expenditure on data center infrastructure is expected to exceed $1.7 trillion by 2030, consulting firm McKinsey estimates, largely because of the expansion of AI. Nvidia has a near-monopoly in supplying chips for data centers and has partnered with companies ranging from OpenAI to Microsoft. "This is a strong endorsement of Nokia's capabilities," said PP Foresight analyst Paolo Pescatore, adding: "Next-generation networks, such as 6G, will play a significant role in enabling new AI-powered experiences". Nokia's shares closed up 20.86% after the deal announcement, a high last touched in late January 2016. PROPOSED PARTNERSHIP MUTUALLY BENEFICIAL, ANALYST SAYS Nokia, which uses Marvell chips for many of its products, and its Swedish rival Ericsson both make equipment for connectivity inside data centers and between data centers and have been benefiting from increased AI use. While Nvidia chips are likely to be more expensive, said Mads Rosendal, analyst at Danske Bank Credit Research, the proposed partnership would be mutually beneficial, given Nvidia's large share in the U.S. data center market. NEW NOKIA CEO BETS ON AI GROWTH, DATA CENTER BUSINESS  Since joining Nokia this year, CEO Justin Hotard, who earlier led Intel's data centers and AI group, has focused on growing its data center business. Nokia's third-quarter profit beat market expectations last week, helped by sales to AI-driven data centers following its acquisition of U.S. optical networking firm Infinera.     Nokia said it will issue 166,389,351 new shares for Nvidia, which the U.S. company will subscribe to for $6.01 per share. Separately, the two companies will also partner with T-Mobile U.S. to develop AI radio technologies for developing 6G and start trials from next year, Nokia said. (Reporting by Supantha Mukherjee, Louise Rasmussen and Johan Ahlander, additional reporting by Gianluca Lo Nostro; Editing by Gareth Jones, Joe Bavier and Alexander Smith)

(The article has been published through a syndicated feed. Except for the headline, the content has been published verbatim. Liability lies with original publisher.)