विज्ञान

(Reuters) -Intel is in early talks to add AMD as a foundry customer, Semafor reported on Wednesday, citing people familiar with the matter. AMD did not immediately respond to a request for comment, and Intel declined to comment. Semafor did not report details of any possible deal. Reuters reported in April that AMD was evaluating Intel's manufacturing technology. (Reporting by Juby Babu in Mexico City and Stephen Nellis in San Francisco; Editing by Alan Barona and Chris Reese)

(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) -Intel is in early talks to add AMD as a foundry customer, Semafor reported on Wednesday, citing people familiar with the matter. (Reporting by Juby Babu in Mexico City; 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 Arasu Kannagi Basil (Reuters) -Fermi's shares opened 19% above the offer price in their Nasdaq debut on Wednesday, valuing the data center real estate investment trust at $14.8 billion amid surging investor enthusiasm for AI infrastructure stocks. The Amarillo, Texas-based company's stock opened at $25, compared with the offer price of $21. While development-stage firms have gone public before, its unusual for a company under a year old with no revenue to pursue an IPO at a $10 billion-plus valuation. "It (Fermi's IPO) speaks to the gold rush happening in AI infrastructure right now. It's a cash geyser," said Matt Kennedy, senior strategist at Renaissance Capital, a provider of IPO-focused research and ETFs. Fermi sold 32.5 million shares at $21 each in an upsized IPO, after marketing them between $18 and $22. Founded in January 2025, Fermi secured a $12.5 billion valuation in the IPO, just months after issuing convertible bonds at a valuation of $3 billion. "Demand for AI-driven infrastructure — and Fermi's plan to pair it with its own modern power sources — has investors intrigued," said Troy Hooper, co-head of equity capital markets, Americas, at Mergermarket. "The company's political ties are seen as another tailwind." Co-founded by former Texas Governor and U.S. Energy Secretary Rick Perry, Fermi has bold ambitions. The company is aiming to build the world's largest energy and data complex, powered by nuclear, natural gas and solar. BUILD IT, THEY WILL COME Project Matador, Fermi's flagship project, is expected to deliver 1.1 gigawatt of power by 2026 end and up to 11 GW at scale. A gigawatt of power is roughly half the Hoover Dam's full generating capacity, according to the U.S. Department of Energy. Last month, Fermi entered a non-binding letter of intent with its first tenant for a 20-year agreement. Still, it does not expect tenant revenue to commence until 2027. "Any investor buying Fermi will need to have a mindset of "if you build it, they will come," Kennedy said. Analysts said execution will be key to sustain market enthusiasm, with investors expected to closely scrutinize signed contracts and progress on its ambitious buildout. A successful listing could pave the way for other long-duration, capital-intensive AI infrastructure bets, according to Hooper. Still, some caution persists. "History offers a cautionary note: many pre-revenue IPOs during the Dot-com era and the recent SPAC boom failed to live up to the hype and ended poorly," Hooper said. (Reporting by Arasu Kannagi Basil in Bengaluru; Editing by Arun Koyyur)

(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 and Max A. Cherney SAN FRANCISCO (Reuters) -Qualcomm has shifted its flagship chips to the latest generation of Arm Holdings' computing architecture with new features aimed at better AI performance, sources familiar with the matter said, a move that could boost Arm's revenues and help Qualcomm's chips compete against offerings from MediaTek and Apple. The decision at least partially settles a question of whether and how Qualcomm would continue to work with Arm technology, following an acrimonious legal battle between the two last year, and the move is likely to raise revenue for Arm which charges more for the newer tech. Arm shares jumped 5% during the regular session after Reuters published the news of Qualcomm's decision to use Arm's latest technology. San Diego, California-based Qualcomm last week introduced a new generation of PC and phone chips. Unlike previous generations, two sources familiar with the matter said Qualcomm's new chips will use the ninth version of Arm's computing architecture, called "v9" in the industry, which includes several improvements aimed at helping chips handle tasks from chatbots and image generators better. Qualcomm rivals such as MediaTek, which has at times eclipsed Qualcomm's market share for mobile phone chips, has publicly confirmed its use of Arm's v9, and most analysts believe Apple uses it, too. The technology is called an instruction set architecture, a fundamental technology that controls what kind of apps can run on a central processing unit. Qualcomm declined to comment on what technology it used for its newest chips but said in a statement that "we chose the instructions that make sense for our customers. That's the beauty of having our own CPU design team – we can pick and choose the instructions that add value." Arm declined to comment.  While Arm faces competition from newer rivals such as RISC-V, an open chip standard that is free to use, that technology is decades less mature than Arm's and has a smaller software developer base. Qualcomm also could have elected to stick with a previous generation of Arm technology, as it did with chips announced last year. How much Qualcomm's move could bolster Arm revenue is difficult to calculate because Qualcomm licenses Arm's computing architecture but designs much of its chip itself rather than buying design elements off the shelf from Arm, said Jay Goldberg, senior analyst for semiconductors and electronics at Seaport Research Partners. But the move is remarkable because of the legal battle, which has not completely ended, Goldberg said. Less than a year ago, Arm had threatened to cancel a key license to Qualcomm, though it later rescinded that threat. “That’s very positive for Arm," Goldberg said of Qualcomm's reported move. “These are companies that were fighting each other. Qualcomm could have gone a very different path here.” (Reporting by Stephen Nellis and Max Cherney in San Francisco; Editing by Stephen Coates)

(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) -Microsoft on Wednesday unveiled Microsoft 365 Premium for individuals at $19.99 a month that bundles the company's Copilot artificial intelligence assistant across apps including Outlook, Excel and Word. The launch marks a shift to simplify and integrate the company's AI offerings more deeply into core products for individual users at a time when technology firms are racing to monetize their investments in advanced AI tools. The company said the plan includes its highest Copilot usage limits and exclusive features such as Researcher, Analyst and Actions, combined with one-terabyte of cloud storage and Microsoft Defender advanced security. The company will stop selling Copilot Pro and existing Copilot Pro and Microsoft 365 Personal or Family customers can switch to Premium. Microsoft, which also has a free version of Copilot, had launched its Copilot Pro at $20 per month last year. OpenAI also offers its ChatGPT Plus for $20 per month. Microsoft added that Personal and Family subscribers will get higher Copilot usage limits to select Copilot features at no extra cost. (Reporting by Jaspreet Singh in Bengaluru; Editing by Shailesh Kuber)

(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 Arasu Kannagi Basil, Echo Wang and Timothy Gardner (Reuters) -Fermi's shares opened 19% above the offer price in their Nasdaq debut on Wednesday, valuing the data center real estate investment trust at $14.8 billion amid surging investor enthusiasm for AI infrastructure stocks. The Amarillo, Texas-based company's stock opened at $25, compared with the offer price of $21. While development-stage firms have gone public before, it's unusual for a company under a year old with no revenue to pursue an IPO at a $10 billion-plus valuation. "It (Fermi's IPO) speaks to the gold rush happening in AI infrastructure right now. It's a cash geyser," said Matt Kennedy, senior strategist at Renaissance Capital, a provider of IPO-focused research and ETFs. Fermi sold 32.5 million shares at $21 each in an upsized IPO, after marketing them between $18 and $22. Co-founded by former Texas Governor and U.S. Energy Secretary Rick Perry, Fermi has bold ambitions. The company is aiming to build the world's largest energy and data complex, powered by nuclear, natural gas and solar. “Having been the Chief Executive Officer for the State of Texas for 14 plus years, I saw a lot of different projects… (but) I’ve never been involved with anything that was more intriguing, from the standpoint of the idea of winning this race towards AI,” Perry told Reuters in an interview. “This is our modern-day Manhattan Project.” Founded in January 2025, Fermi secured a $12.5 billion valuation in the IPO, just months after issuing convertible bonds at a valuation of $3 billion. "Demand for AI-driven infrastructure — and Fermi's plan to pair it with its own modern power sources — has investors intrigued," said Troy Hooper, co-head of equity capital markets, Americas, at Mergermarket. "The company's political ties are seen as another tailwind." BUILD IT, THEY WILL COME Project Matador, Fermi's flagship project, is expected to deliver 1.1 gigawatt of power by 2026 end and up to 11 GW at scale. A gigawatt of power is roughly half the Hoover Dam's full generating capacity, according to the U.S. Department of Energy. Last month, Fermi entered a non-binding letter of intent with its first tenant for a 20-year agreement. Still, it does not expect tenant revenue to commence until 2027. "Any investor buying Fermi will need to have a mindset of "if you build it, they will come," Kennedy said. Analysts said execution will be key to sustain market enthusiasm, with investors expected to closely scrutinize signed contracts and progress on its ambitious buildout. A successful listing could pave the way for other long-duration, capital-intensive AI infrastructure bets, according to Hooper. Still, some caution persists. "History offers a cautionary note: many pre-revenue IPOs during the Dot-com era and the recent SPAC boom failed to live up to the hype and ended poorly," Hooper said. (Reporting by Arasu Kannagi Basil in Bengaluru, Echo Wang in New York and Timothy Gardner in Washington, D.C.; Editing by Arun Koyyur)

(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) -Microsoft Chief Commercial Officer Judson Althoff will take on an expanded role as CEO of commercial business, freeing up company chief Satya Nadella to focus more on the technical side of the business in the tech giant's quest to dominate the AI race. Althoff will lead what Microsoft CEO Nadella called a new organization, which will combine sales, marketing and operations. Nadella said the reorganization would help him and other engineering leaders to be "laser focused on our highest ambition technical work – across our datacenter buildout, systems architecture, AI science, and product innovation," in a blog post on Wednesday. Althoff will also be the chief of a new commercial leadership team that includes leaders from engineering, sales, marketing, operations and finance. "We are in the midst of a tectonic AI platform shift, one that requires us to both manage and grow our at-scale commercial business today, while building the new frontier and executing flawlessly across both," Nadella said. In 2021, Microsoft brought together its global sales and marketing organization and worldwide commercial business into a single unified organization, helmed by Althoff, who joined the company in 2013 as president of Microsoft North America. Microsoft last month said it is combining the separate marketplaces for its AI tools aimed at businesses into one offering called "Microsoft Marketplace." It previously offered tools for software developers, who use its Azure cloud computing service, on one marketplace, and applications and so-called "agents" — AI tools designed to carry out tasks on behalf of human users within applications — on another. (Reporting by Juby Babu in Mexico City; Editing by Sahal Muhammed)

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

Personal impressions often determine your chances at a promotion or recognition in the workplace. If you're mostly working from home, how can you stay relevant and help keep your career on track? Berlin (dpa) – Working from home has become routine for many of us, particularly since the coronavirus pandemic. There are perks aplenty: You don't have to commute or even change out of your pyjamas, can attend to your family's needs – and the fridge isn't far. On the other hand, you're out of your superior's sight. Yes, they're not around to breathe down your neck, but you could be "out of mind" for promotions and raises. Is there a way to avoid career stagnation while working remotely? Two experts weigh in: How much of your working time should you spend at your home workplace? Not all employers have fixed rules on when and how often employees have to come to the office. Given the choice, employees often prefer to work from home. From a career perspective, this isn't always advisable though, says Bastian Hughes, a career coach and host of a career advice podcast. So long as your job isn't supposed to be entirely done remotely, he recommends showing up at the office regularly. Depending on your individual needs and any arrangements you have with your co-workers, he suggests that it be about once a week or even once a month. This allows you to make or deepen personal contacts, which can be advantageous for your career development.  Video conferencing can't always make up for personal interactions. "Meeting face to face promotes trust, a feeling of belonging and empathy," remarks career advisor Ragnhild Struss. In a "world of digital sensory overload and artificial intelligence," maintaining human presence is becoming the new currency, she says. "This is one of the reasons that employees who also show up on-site are more likely to be promoted, since their superiors feel more comfortable with them." How can you stay visible while working from home? First of all, here's something that should be self-evident: When there's communication via an online team collaboration platform, always turn on your camera so that you're visible during the discussions or meetings," Hughes says. It's also important to be proactive – for example to communicate, for everyone to see, what you're working on, your results, and where you need support.    Another way to stay visible is to arrange short, informal talks. "You should use these conversations with co-workers and superiors not only for the purpose of coordination, but for relationship management as well," Struss says. Warmly passing along your appreciation to a co-worker for something they've done – naming them and specifying the reason – can also be helpful, according to Struss. By making others more visible, you become more visible yourself. Make it clear what it is that motivates you by talking about it on occasion. "People primarily follow attitudes that impress them," Struss says. This, too, helps you to stay visible. Network with people outside your immediate work team. It can be useful, for instance, to initiate an exchange of ideas with co-workers from other departments. "Not least of all, it's important to show interest in the big picture, beyond your own concerns," says Struss. What other guidelines should you follow? Employees who split their working time between their home and the office should clearly communicate when they can be reached. Hughes recommends, for example, that you keep your availability status up-to-date using a collaboration platform such as Microsoft Teams or Slack. You shouldn't be permanently available, however. "Being visible doesn't mean always being online," remarks Struss, who says true presence comes from being focused. This means protecting your deep work time, scheduling breaks and setting limits, she explains. As Struss sees it, an employee who regularly, reliably delivers, both on a professional as well as on a personal level, will be regarded as a stable factor in the team. "Reliability, particularly for someone working remotely, is the strongest form of visibility," she says. How do you, as a remote worker, avoid the urge to appear especially busy? Some remote workers feel a need to show they're very busy by immediately answering every email, constantly creating new projects, filling their calendar with numerous meetings, etc. But this isn't necessary, according to Struss, noting that not every work lull is a sign of inefficiency. "You need to be confident in your performance and in your worth," she says. It helps to ask yourself, "What's really important right now?" And, "What's part of my job, and what's simply pressure that I'm putting on myself?" "You should show, in conversations or in meetings, what you're accomplishing, rather than try to appear that you're more or less always sitting at your computer," Hughes says. How best to speak with your superior? If you feel you've fallen through the cracks by working from home, you might want to bring it up with your superior. Simply emailing them isn't advisable, in Hughes' view. It's better to request a face-to-face meeting during which you dispassionately voice your fear, along with your wishes and expectations. It's important to share your sense of things in a non-accusatory way. Struss recommends pointing to concrete examples, say, projects you feel you were left out on. A constructive approach is helpful, she says, namely by asking your superior how you can jointly ensure that your skills are put to better use. The following information is not intended for publication dpa/tmn eut yyzz a3 bzl amb amc ob

(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) -Swedish self-driving truck startup Einride has raised around $100 million in the latest funding round from investors, including asset management firm EQT Ventures and quantum computing company IonQ, at an undisclosed valuation. The company on Wednesday said the capital will allow it to scale the deployment of its autonomous freight trucks, further technology development and expand its customer base. Autonomous trucking technology promises to grow the freight and long-haul industry, but faces regulatory scrutiny over self-driving tech. Self-driving freight services require less mapping than other technologies as they run on fixed routes between predefined points, mostly on major highways without intersections or pedestrians. In 2021, Einride raised $110 million from investors such as Maersk's venture capital arm and Singapore state investor Temasek. Later that year, the company expanded into the U.S. market and signed up GE Appliances, Swedish vegan milk maker Oatly and tiremaker Bridgestone as customers. (Reporting by Zaheer Kachwala in Bengaluru; Editing by Vijay Kishore)

(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) -Stargate, a $500 billion AI infrastructure initiative that is central to Washington's push for dominance in the field, has drawn a slew of investors and suppliers since its launch in January. The project will be developed by ChatGPT parent OpenAI, SoftBank and Oracle and is intended to generate 10 gigawatts in total data center capacity. Here's a list of major partners on the Stargate project: ORACLE Oracle will open three new data center sites with OpenAI in Shackelford County, Texas; Dona Ana County, New Mexico; and an undisclosed site in the Midwest as part of efforts to expand Stargate's compute capacity. Oracle and OpenAI announced in July plans to develop up to 4.5 gigawatts of U.S. data center capacity on top of what the two companies were already building at the first Stargate facility in Abilene, Texas. SOFTBANK SoftBank is the lead equity investor and financial backer for the Stargate project. OPENAI The ChatGPT maker's role in the Stargate project includes planning, site selection and overseeing the deployment of advanced computing capacity. At the time of the project's announcement, OpenAI committed to deploy $100 billion immediately. NVIDIA The AI chip leader is among the key partners for Stargate, providing its top-of-the-line processors for the project. Stargate's Norway chapter will initially install 100,000 Nvidia chips and could see a tenfold expansion in future phases. The Stargate UAE project will use Nvidia's Grace Blackwell GB300 systems. Nvidia has also said it would invest up to $100 billion in OpenAI and supply it with data center chips. COREWEAVE The AI cloud provider is among the providers for the massive data center capacity Stargate requires. Ongoing projects with CoreWeave and Oracle have contributed to bringing Stargate's planned capacity to nearly 7 gigawatts, OpenAI said late in September. OpenAI and CoreWeave announced a new deal worth up to $6.5 billion last month, bringing the total value of their agreements to $22.4 billion. SAMSUNG and SK HYNIX The South Korean companies have signed letters of intent to supply memory chips for the project. South Korea's top presidential adviser, Kim Yong-beom, said on Wednesday OpenAI was seeking to order 900,000 semiconductor wafers in 2029, and it planned to set up joint ventures with Samsung and SK Hynix to build two data centers in South Korea with an initial capacity of 20 megawatts. ARM HOLDINGS The chip architecture provider is a "technology partner" on Stargate. CEO Rene Haas has coined Arm's inclusion in the project, over the x86 technology used by rivals Intel and Advanced Micro Devices, as a sign of the company's importance to the AI ecosystem. G42, CISCO State-backed UAE firm G42 will build the "Stargate UAE" project as part of a deal brokered earlier this year by U.S. President Donald Trump to build the world's largest set of AI data centers outside the United States. Cisco Systems is also a partner on the project, along with other Stargate backers. MGX United Arab Emirates firm MGX is among the investors for the venture. CRUSOE "Noecloud" Crusoe was tapped by Oracle to build some data centers for Stargate, Reuters has reported. Crusoe provides specialized AI cloud and data-center services. NSCALE GLOBAL HOLDINGS AND AKER ASA OpenAI is launching its first Stargate data centre project in Europe, partnering with developer Nscale Global Holdings and investment group Aker ASA to open a $1-billion facility in Norway next year. Nscale and Aker will jointly own Stargate Norway with 50% stakes each, investing about $1 billion for the initial phase of the project, which will consume about 20 megawatts of electrical power capacity and have OpenAI as its first customer. FOXCONN Foxonn has said it plans to manufacture data centre equipment with SoftBank at the Taiwanese firm's former electric vehicle factory in Ohio as part of the Stargate project. (Reporting by Arsheeya Bajwa and Harshita Mary Varghese in Bengaluru; 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.)