विज्ञान

(Reuters) -Nvidia CEO Jensen Huang said on Tuesday he expects to make announcements with South Korean companies that will please U.S. President Donald Trump and South Korea when he visits the country this week. Huang plans to meet global leaders and top Korean executives when he attends the Asia-Pacific Economic Cooperation CEO Summit in South Korea, the U.S. artificial intelligence chip giant said earlier. "Samsung, SK, Hyundai, LG, Naver, if you just look across the entire Korean ecosystem, every one of the companies are deep friends of mine and very good partners," he told reporters on the sidelines of the Nvidia developers' conference held in Washington on Tuesday. "And so when I go hopefully we'll have some announcements that will be really, really delightful to the people of Korea and really delightful to President Trump, but I'll save it for a few more days," he said. He said Nvidia is partnering with Samsung Electronics and Hyundai "in many ways," saying the two companies are "going to invest in AI factories." Trump is due to address a summit of CEOs and meet with South Korean President Lee Jae Myung in Gyeongju on Wednesday. (Reporting by Hyunjoo Jin in Seoul and Stephen Nellis in Washington; Editing by Sonali Paul)

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

SEOUL (Reuters) -South Korea's SK Hynix posted on Wednesday a record quarterly profit, spurred by an artificial intelligence boom that has driven up demand for both advanced chips and conventional products. The Nvidia supplier reported an 11.4 trillion won ($8.02 billion) operating profit for the July-September period, up 62% from a year earlier. The figure was in line with an 11.4 trillion won forecast by LSEG SmartEstimate, which is weighted toward analysts who are more consistently accurate. Quarterly revenue for the South Korean firm rose 39% in the third quarter to 24.4 trillion won. ($1 = 1,421.4800 won) (Reporting by Heekyong Yang, Hyunjoo Jin and Joyce Lee; Editing by Leslie Adler and Jamie Freed)

(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 Stephen Nellis WASHINGTON (Reuters) -Nvidia and Palantir Tech on Tuesday announced a deal in which Palantir will tap Nvidia's chips and software to help its customers speed up decision making in complex fields such as logistics. Well-known as a defense contractor, Palantir's shares have risen this year, driven by increased defense spending on its military-grade AI tools. But the Denver-based company is also pursuing corporate customers, where its platform can ingest data from many different corporate systems, such as staffing and inventory software, to give executives a real-time view of business performance. Under the deal struck Tuesday and announced at a conference in Washington, Nvidia's software will be available on Palantir's platforms. The two said they would help companies with logistics problems, such as when shipments of products from Asia to the U.S. are delayed by storms. Nvidia technology can generate potential new shipping routes around the storms and its AI agents can then compare multiple options based on product costs and customer demand in different regions to help select the best one. "Because of the speed that you do these optimizations, you can then run them hourly to re-optimize your supply chain and rethink changes," Justin Boitano, vice president of enterprise AI at Nvidia, told Reuters in an interview. The companies did not disclose the financial terms of the deal. Kevin Kawasaki, global head of business development at Palantir, said that AI systems on their own can struggle to understand vast streams of corporate data. Palantir's systems organize the data streams to reflect business conditions in a way that both humans and AI systems powered by Nvidia can use. "It's got to be tethered in reality," Kawasaki told Reuters, referring to how AI systems can help in decision making. "It's not overly predictive, so much as it's just it's very good at recognizing the present. And that's when you can start making better next-second decisions." (Reporting by Stephen Nellis in Washington; Editing by Sam Holmes)

(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 Blake Brittain (Reuters) -The maker of celebrity video platform Cameo sued OpenAI in a California federal court on Tuesday, arguing that the new "Cameo" feature of OpenAI's Sora video generation app violates its trademark rights. Cameo said in the complaint that OpenAI's Cameo, which allows users to create and share their virtual likenesses on Sora, is likely to cause consumer confusion and dilute its brand. An OpenAI spokesperson said the company is reviewing the complaint but "disagree[s] that anyone can claim exclusive ownership over the word 'cameo.'" Cameo CEO Steven Galanis said in a statement that the company had tried to resolve the dispute "amicably" but OpenAI refused to stop using the Cameo name. Cameo, founded in 2017, allows users to hire celebrities to appear in short, personalized videos. Artificial intelligence giant OpenAI's Sora, launched as a standalone app on September 30, lets users create and share AI-generated videos. OpenAI calls Sora users' likenesses on the app "Cameos." Cameo said OpenAI chose the name "in blatant disregard for the obvious confusion it would create." Cameo also said that Sora allows users to create videos with celebrity "Cameos" including entrepreneur Mark Cuban and boxer and influencer Jake Paul, putting the AI giant in direct competition with Cameo's platform. "Users seeking a personalized celebrity video now have a choice – use Plaintiff's CAMEO service to book talent and receive an authentic, custom video prepared by that celebrity, or use Sora’s 'Cameo' service to create an extremely realistic AI-generated video featuring a celebrity’s likeness," the lawsuit said. Cameo requested an unspecified amount of monetary damages and a court order blocking OpenAI from using the "Cameo" name. (Reporting by Blake Brittain in Washington; Editing by Sam Holmes and Christopher Cushing)

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

SEOUL (Reuters) -South Korea's SK Hynix gave a bullish outlook and said it would boost investments next year after posting a record quarterly profit on Wednesday, spurred by an AI boom that has stoked demand for both advanced chips and conventional products. SK Hynix said it has already secured full customer demand for its entire DRAM and NAND production for next year, while planning to start shipping the next-generation HBM4 chips in the fourth quarter of this year. While thriving on a strong market for conventional memory chips, the company wants to maintain its edge in advanced AI chips supplied to Nvidia, in the face of mounting competition from rivals such as Samsung Electronics and Micron. The Nvidia supplier reported an 11.4 trillion won ($8.02 billion) operating profit for the July-September period, up 62% from a year earlier and in line with a forecast by LSEG SmartEstimate. "Demand across the memory segment has soared due to customers' expanding investments in AI infrastructure," SK Hynix said in a statement. SK Hynix expects DRAM shipments to grow by more than 20% year-on-year in 2026. The chip industry's shift to producing high-performance chips, driven by the spending binge on artificial intelligence, has constrained capacity for traditional chips, raising the prospect of a shortage and an extended rise in memory prices, analysts said. Quarterly revenue for the South Korean firm rose 39% in the third quarter to 24.4 trillion won. SK Hynix said it had completed discussions with key customers regarding high-bandwidth memory (HBM) supply for next year, and plans a full-scale sales expansion of HBM4 products next year. Buoyed by its advantage in advanced chips, SK Hynix's shares have surged 200% so far this year, outpacing an 87% gain in Samsung's shares and a 67% gain in the benchmark KOSPI. ($1 = 1,421.4800 won) (Reporting by Heekyong Yang, Hyunjoo Jin and Joyce Lee; Editing by Jamie Freed and Sonali Paul)

(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 Stephen Nellis, Alexandra Alper and Arsheeya Bajwa WASHINGTON (Reuters) -Nvidia CEO Jensen Huang came to Washington on Tuesday with a message for the Trump administration: the U.S. can win the AI battle if the world, including China's massive developer base, runs on Nvidia systems.  In his address at the first Nvidia developers' conference held in Washington, Huang walked a fine line between praising President Donald Trump, whose "America First" agenda Huang credited with spurring greater investment in U.S. manufacturing and AI leadership, and risking further antagonism of China. Huang said the artificial intelligence chip leader will build seven new supercomputers for the U.S. Department of Energy and had $500 billion in bookings for advanced chips, but also lamented that the Chinese government has shut it out of its market. Nvidia is at the core of the global AI rollout, and it is striking deals around the world while also navigating a U.S.-China trade war that could determine which country's technology is most used around the world. Trump is touring Asia ahead of an expected meeting with Chinese President Xi Jinping on Thursday where China's use of Nvidia's chips could be a key issue. "We want America to win this AI race. No doubt about that," Huang said. "We want the world to be built on American tech stack. Absolutely the case. But we also need to be in China to win their developers. A policy that causes America to lose half of the world's AI developers is not beneficial long term, it hurts us more."  Huang said his company had not applied for U.S. export licenses to send its newest chips to China because of the Chinese position. "They've made it very clear that they don't want Nvidia to be there right now," he said at a news conference during the company's GTC developers event. "I hope that will change in the future because I think China is a very important market." U.S. administrations have swung back and forth on allowing Nvidia's advanced chips into China, vacillating on whether access would make China more dependent on the U.S. technology or give its military and tech companies a competitive boost. Huang praised Trump while announcing new products and deals. These included network technology that will let Nvidia AI chips work with quantum computers, a telecom deal with Finland's Nokia and self-driving car technology with Uber and Stellantis. The supercomputers Nvidia is building for the Energy Department will in part help the United States maintain and develop its nuclear weapons arsenal. The supercomputers will also be used to research alternative energy sources such as nuclear fusion. The largest of the supercomputers for the Department of Energy will be built with Oracle and contain 100,000 of Nvidia's Blackwell chips. "Putting the weight of the nation behind pro-energy growth completely changed the game," Huang said. "If this didn't happen, we could have been in a bad situation, and I want to thank President Trump for that." Nvidia shares closed 5% higher at $201.03 on Tuesday. Blake Anderson, an associate portfolio manager at Carson Group,  estimated that one of the supercomputers, dubbed "Solstice," could alone feature Nvidia chips worth about $3 billion to $4 billion.  However, since federal customers are likely to receive discounts, the pricing could vary from the $30,000 to $40,000 price tag Blackwell chips tend to carry, Anderson added. Nvidia declined to comment on the size of the Energy Department deals, and the Energy Department did not immediately respond to requests for comment.  Among the deals Huang discussed, Nvidia and Nokia will target the AI communications market. Nvidia will invest $1 billion for a 2.9% stake in Nokia and it also introduced a new product line called Arc, designed to work with telecommunications equipment. Huang said Nvidia will work with Nokia to improve the power efficiency of the company's base stations for 6G, the next generation of wireless data technology. “We're going to take this new technology and we’ll be able to upgrade millions of base stations around the world,” Huang said. Nvidia also announced a partnership with Palantir Technologies, a company that works closely with the U.S. government, to improve logistics for companies.  Nvidia announced a new self-driving car technology platform called Hyperion. Huang said that Nvidia is partnering with Uber to create a network of robotaxis.  "These announcements all show Nvidia's ability to extend its reach beyond its core data center customers," said Gil Luria, an analyst at D.A. Davidson. "While these projects pale in comparison to the capex by the hyperscalers such as Microsoft, Amazon, Google and Meta, they could create new markets for Nvidia down the line." Trump in his second term at first restricted exports of Nvidia's AI chips designed for the China market before reversing course in July.  Huang has argued that Nvidia needs access to some $50 billion in potential sales from the Chinese market to fund U.S.-based research and development to maintain his company's edge. Reuters has previously reported that Chinese developers still want Nvidia's chips, despite pressure from Beijing to purchase domestic chips from Huawei Technologies Co. Nvidia outlined how it is making chips in Arizona at TSMC's facilities, and assembling servers in Texas and networking gear in California. “We are manufacturing in America again – it is incredible. The first thing that President Trump asked me is, 'bring manufacturing back,'" Huang said. He said that TSMC is bringing its most advanced technology for packaging chips to the United States in the next several months. (Reporting by Stephen Nellis and Alexandra Alper in Washington; Additional reporting by Arsheeya Bajwa in Bengaluru; Editing by Peter Henderson and Matthew Lewis)

(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 Blake Brittain (Reuters) -The maker of celebrity video platform Cameo sued OpenAI in a California federal court on Tuesday, arguing that the new "Cameo" feature of OpenAI's Sora video generation app violates its trademark rights. Cameo said in the complaint that OpenAI's Cameo, which allows users to create and share their virtual likenesses on Sora, is likely to cause consumer confusion and dilute its brand. An OpenAI spokesperson said the company is reviewing the complaint but "disagree[s] that anyone can claim exclusive ownership over the word 'cameo.'" Cameo CEO Steven Galanis said in a statement that the company had tried to resolve the dispute "amicably" but OpenAI refused to stop using the Cameo name. Cameo, founded in 2017, allows users to hire celebrities to appear in short, personalized videos. Artificial intelligence giant OpenAI's Sora, launched as a standalone app on September 30, lets users create and share AI-generated videos. OpenAI calls Sora users' likenesses on the app "Cameos." Cameo said OpenAI chose the name "in blatant disregard for the obvious confusion it would create." Cameo also said that Sora allows users to create videos with celebrity "Cameos" including entrepreneur Mark Cuban and boxer and influencer Jake Paul, putting the AI giant in direct competition with Cameo's platform. "Users seeking a personalized celebrity video now have a choice – use Plaintiff's CAMEO service to book talent and receive an authentic, custom video prepared by that celebrity, or use Sora’s 'Cameo' service to create an extremely realistic AI-generated video featuring a celebrity’s likeness," the lawsuit said. Cameo requested an unspecified amount of monetary damages and a court order blocking OpenAI from using the "Cameo" name. (Reporting by Blake Brittain in Washington; Editing by Sam Holmes and Christopher Cushing)

(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. The restructuring has been one of many initiatives under way at OpenAI, which is betting that its generative AI technology will ultimately transform the global economy. The moves have heartened OpenAI's investors and business partners, including Nvidia's CEO Jensen Huang, who was asked about the news at a developer conference on Tuesday.  "I wish back in the earlier days that we had invested a lot more," Huang said. "If you told me that OpenAI is going to go public next year, I'm not surprised, and in a lot of ways, I think this can be one of the most successful public offerings in history."  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 roughly 30 gigawatts of data-center infrastructure over the next few years.  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 artificial general intelligence, and the new deal requires an independent panel to verify OpenAI's claims that it has reached AGI, the point at which AI systems can match a well-educated human adult. 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%, 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 AGI. 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.)

SEOUL (Reuters) -South Korea's SK Hynix posted on Wednesday a record quarterly profit, spurred by an artificial intelligence boom that has driven up demand for both advanced chips and conventional products. The Nvidia supplier reported an 11.4 trillion won ($8.02 billion) operating profit for the July-September period, up 62% from a year earlier. The figure was in line with an 11.4 trillion won forecast by LSEG SmartEstimate, which is weighted toward analysts who are more consistently accurate. Quarterly revenue for the South Korean firm rose 39% in the third quarter to 24.4 trillion won. SK Hynix said it had completed discussions with key customers regarding high-bandwidth memory (HBM) supply for next year. The company added that it planned to begin shipments of its latest HBM4 products in the fourth quarter after completing development in September and entering mass production. A full-scale sales expansion of HBM4 products is planned for next year, SK Hynix said in a statement. ($1 = 1,421.4800 won) (Reporting by Heekyong Yang, Hyunjoo Jin and Joyce Lee; Editing by Leslie Adler and Jamie Freed)

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

BRASILIA (Reuters) -Some automakers may halt operations in Brazil within two to three weeks if the global chips supply crisis continues, an official said on Tuesday, adding the Brazilian government is contacting Chinese authorities to find a solution. Uallace Moreira, a secretary in Brazil's Development, Industry, Trade and Services Ministry, did not specify which automakers in Latin America's largest economy could be hit first by the global crisis, triggered by a stand-off between China and the Netherlands over chipmaker Nexperia. "If there is no solution in this short period of time, in two or three weeks, there may be a shutdown for some automakers," Moreira told journalists after a meeting with the president of Brazilian automakers association Anfavea. Brazil's Vice President Geraldo Alckmin, who also heads the ministry, attended the meeting as well. In September, the Dutch government took control of Nexperia, which manufactures chips for cars and consumer electronics, as it cited worries about the possible transfer of technology to the firm's Chinese parent company, Wingtech. In response, the Chinese government blocked Nexperia from exporting its products from China. While most of the company's chips are produced in Europe, around 70% are packaged in China before distribution. Moreira noted Alckmin has already contacted both the Brazilian ambassador to China and the Chinese ambassador to Brazil to initiate discussions about the matter as the local private sector requests dialogue between the parties. (Reporting by Victor Borges in Brasilia; Writing by Andre Romani and Fernando Cardoso; Editing by Kylie Madry and Aurora Ellis)

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