Inkhabar webdesk

(Reuters) -Anthropic is expanding its deal with Google to use as many as one million of the tech giant's artificial intelligence chips, worth tens of billions of dollars, as the startup races to advance its AI systems in the competitive market. Under the deal announced on Thursday, Anthropic will have access to more than one gigawatt of computing capacity, coming online in 2026, to train the next generations of its Claude AI model on Google's in-house tensor processing units, or TPUs, which were traditionally reserved for internal use. Anthropic said it chose the TPUs due to their price-performance ratio and efficiency, as well as its existing experience in training and serving its Claude models with the processors. The deal is the latest sign of insatiable chip demand in the AI industry, where companies are rushing to develop technology that can match or surpass human intelligence. Alphabet-owned Google, whose TPUs are available for rent on Google Cloud and serve an alternative to supply-constrained Nvidia chips, will also provide additional cloud computing services to Anthropic. Rival OpenAI recently signed multiple deals that may cost over $1 trillion to secure about 26 gigawatts of computing capacity, enough to power roughly 20 million U.S. homes. One gigawatt of compute can cost roughly $50 billion, industry executives have said. ChatGPT-maker OpenAI is actively using Nvidia's graphics processing units and AMD's AI chips to power its growing demand. Reuters exclusively reported earlier in October that Anthropic is projecting to more than double, and potentially nearly triple, its annualized revenue run rate next year, fueled by the rapid adoption of its enterprise products. The startup emphasizes AI safety and building models for enterprise use cases. Its models have helped power a boom in vibe coding startups such as Cursor. (Reporting by Juby Babu in Mexico City and Zaheer Kachwala in Bengaluru; Editing by Shilpi Majumdar, Anil D'Silva and 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 Arsheeya Bajwa, Max A. Cherney and Stephen Nellis (Reuters) -Intel beat expectations for September-quarter profit as CEO Lip-Bu Tan's drastic cost-cutting measures helped the chipmaker shore up its finances amid a slew of high-profile investments in the company. Shares were up 7% in after-hours trading.   This marks the Santa Clara, California-based company's first earnings announcement after multibillion-dollar investments from Nvidia and Japan's SoftBank as well as an unprecedented U.S. government stake, with investors anticipating a major cash boost.  Intel received the SoftBank money in the third quarter but has not yet received Nvidia's cash, finance chief Dave Zinsner said in an interview with Reuters. The investments are expected to be a major lifeline for Intel, which has been struggling to maintain its dominance in the PC and server central processing unit markets as it competes with Advanced Micro Devices, while repeated attempts to break into the AI chip market commanded by Nvidia have not produced results.  After its share price dropped about 60% last year due to these concerns, Intel stock has risen nearly 90% so far in 2025 thanks to the new investments in the company, helping it outperform the stock of artificial-intelligence chip leader and Wall Street darling Nvidia.  "Shares popped after-hours based on better-than-feared guidance ex-Altera, visible cost and gross margin progress, AI-PC buzz, and $15B of fresh strategic funding that shores the balance sheet," said Michael Schulman, chief investment officer at Running Point Capital. Intel acquired Altera in 2015 for $16.7 billion but sold a 51% stake in the programmable chip designer to Silver Lake in September. Nvidia said last month it would invest $5 billion in Intel, giving it a stake of about 4% after new shares are issued. In August, Intel secured $2 billion from SoftBank. After U.S. President Donald Trump called for CEO Tan's resignation over China ties, a hastily arranged Washington meeting produced an unusual deal involving the U.S. government taking a 10% stake for $8.9 billion.  On a conference call with investors, Tan said that a newly formed central engineering group will streamline Intel's work on its own chip designs and will also offer custom-designed chips for external customers, a business where it will compete against Broadcom and Marvell Technologies. Those two companies are both helping customers such as Alphabet's Google and Amazon make their AI chips.  Demand for Intel's chips was strong enough during the third quarter that the company is tight on supply, CFO Zinsner said. One of the factors is that data center operators have realized that they need to upgrade the central processing unit in order to keep pace with the advanced AI chips, he said. Upgrading the CPU gives power and performance advantages that help build and run AI applications. "We're under-shipping demand at this point, which I guess is a high-class problem," Zinsner said. During the conference call, Zinsner said that Intel's yields – a measure of how many of the chips it makes come out functioning well – for its 18A manufacturing process "are not where we need them to be in order to drive the appropriate level of margins." Zinsner said yields would likely not hit "an industry-acceptable level" until 2027.  Reuters reported in August that Intel is struggling with yields on its 18A manufacturing technology. For the September quarter, it recorded adjusted gross margins of 40%, beating estimates of 35.7%, while adjusted profit of 23 cents per share also outperformed expectations of a profit of 1 cent per share, according to data compiled by LSEG.  The company will end the year with a workforce that is over one-fifth smaller than last year, it had said in July, while Tan has significantly pared back the expensive manufacturing ambitions of his ousted predecessor, Pat Gelsinger. Tan has aggressively cut costs and divested assets, after Gelsinger's ambitions of turning Intel into a contract manufacturer of chips that would compete against Taiwan's TSMC led to the company in 2024 recording its first annual loss since 1986. Intel forecast current-quarter revenue between $12.8 billion and $13.8 billion, with a midpoint of $13.3 billion, compared with analysts' average estimate of $13.37 billion. Zinsner said that Intel plans capital expenditures of $27 billion in 2025 versus $17 billion in 2024. (Reporting by Arsheeya Bajwa in Bengaluru and Max A. Cherney and Stephen Nellis in San Francisco; Additional reporting by Aditya Soni in Bengaluru; Editing by Sayantani Ghosh 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.)

(Reuters) -Digital Realty Trust raised its annual forecasts for core funds from operations and revenue on Thursday, as it expects an increase in demand for data center services due to the boom in artificial intelligence. Digital Realty has been benefiting from a surge in demand as more enterprises upgrade and outsource their IT infrastructure, further boosted by advancements in AI technology. Large-scale cloud service providers called hyperscalers are racing to build AI infrastructure, with Amazon, Meta, Alphabet and Microsoft projected to spend over $360 billion in 2025, according to company filings. Most of the investment is expected to go toward powering data centers. The growing demand for data centers is expected to accelerate Digital Realty's leasing activity, thereby aiding revenue growth. Digital Realty leases out managed data centers to clients in sectors that range from cloud and information technology to social networking, communications and manufacturing. The company earlier in October announced a strategic collaboration with Dell Technologies and DXC to bring AI directly to customers' data through a combination of validated use cases and expert-led implementation and end-to-end management. Austin, Texas-based Digital Realty now expects annual core FFO to be in the range of $7.32 to $7.38 per share, compared with its prior outlook of between $7.15 to $7.25. The real estate investment trust now expects revenue between $6.03 billion and $6.08 billion for the year, compared with its prior forecast of between $5.93 billion and $6.03 billion. For the third quarter ended September 30, Digital Realty posted revenue of $1.58 billion, above estimates of $1.53 billion. Core FFO came in at $1.89 per share in the quarter, compared with $1.67 last year. (Reporting by Juby Babu in Mexico City; Editing by Shreya Biswas and Maju Samuel)

(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) -Tesla said on Thursday it is recalling 63,619 Cybertrucks over software that makes the front parking lights too bright, potentially impairing oncoming drivers' vision. It said an over-the-air software update has been released to the fix the issue at no charge for the Cybertrucks built between November 13, 2023 and October 11, 2025. The Elon Musk-led automaker said it identified the problem during an internal review earlier this month after photometric tests confirmed the excessive brightness. Tesla said it has not received any reports of crashes, injuries, or fatalities related to the issue. On Wednesday, the EV maker also recalled 12,963 Model 3 and Model Y vehicles over a battery-pack component defect that could cause a loss of propulsion and increase crash risk. Earlier this month, the U.S. National Highway Traffic Safety Administration said it was opening an investigation into 2.88 million Tesla vehicles equipped with its Full Self-Driving system after more than 50 reports of traffic-safety violations and a series of crashes. The company reported record third-quarter revenue that beat Wall Street estimates on Wednesday, driven by its highest quarterly EV sales as U.S. buyers rushed to lock in a key tax credit before it expired last month. But Tesla's profit missed analysts' expectations, in part due to tariff and research costs, as well as a decline in income from regulatory credits that are expected to continue fading with recent legislation passed by the Trump administration. Shares of the company were down 3.3% in premarket trading. The stock is up nearly 9% so far this year. (Reporting by Jaspreet Singh and Gnaneshwar Rajan in Bengalur; Editing by Mrigank Dhaniwala, Janane Venkatraman and Tasim Zahid)

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

RIO DE JANEIRO (Reuters) -U.S. data center operator Equinix is expanding its operations in Brazil, which the firm sees as a priority market, its managing director for Latin America Eduardo Zago said on Thursday. WHY IT'S IMPORTANT Equinix is one of the companies benefiting from the increased demand for data center infrastructure amid a surge in artificial intelligence usage. Brazil has been seeking to become a sustainable infrastructure hub, given the country's abundant supply of renewable energy. The government has also proposed to exempt data center-related equipment from federal taxes to lure investment. BY THE NUMBERS Equinix has eight data centers in Brazil, with a ninth under construction, Zago said during an event in Rio de Janeiro. All are located across the country's southeastern states of Sao Paulo and Rio de Janeiro, he added. Equinix also bought five real estate areas for a potential future expansion in Brazil, according to Zago. However, the firm does not have a plan for them so far, he said, adding that the company will grow depending on demand. KEY QUOTES "Brazil is a priority region for us," Zago said. "It's a market where there is no lack of space nor money," he added, noting the company is in "strong expansion" in the country. CONTEXT In Latin America, Equinix also has data center operations in Mexico, Chile, Peru and Colombia. (Reporting by Rodrigo Viga Gaier in Rio de Janeiro; Writing by Andre Romani; Editing by 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.)

(Reuters) -AI data centre startup Crusoe is raising $1.38 billion at a valuation of about $10 billion from an anticipated Series E funding round, the Denver-based company said on Thursday The oversubscribed round is being co-led by Valor Equity Partners and Mubadala Capital and includes Nvidia, Fidelity Management and Founders Fund as some of its major investors. The company, which was involved in building OpenAI's first big data centre in the U.S., has raised about $3.9 billion since its inception in 2018. It recently announced that the first phase of its 1.2 gigawatt data center in Abilene, Texas campus, was live just one year after construction began. (Reporting by Anuj T 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.)

(Reuters) -AI data centre startup Crusoe is raising $1.38 billion at a valuation of about $10 billion from an anticipated Series E funding round, the Denver-based company said on Thursday The oversubscribed round is being co-led by Valor Equity Partners and Mubadala Capital and includes Nvidia, Fidelity Management and Founders Fund as some of its major investors. The company, which was involved in building OpenAI's first big data centre in the U.S., has raised about $3.9 billion since its inception in 2018. It recently announced that the first phase of its 1.2 gigawatt data center in Abilene, Texas campus, was live just one year after construction began. (Reporting by Anuj T 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.)

(Reuters) -Applied Materials will cut about 4% of its workforce, or roughly 1,400 jobs, to streamline operations, the chip equipment maker said on Thursday as tighter U.S. export controls on semiconductors weigh on its business. The company, one of the largest U.S. makers of semiconductor manufacturing equipment, will take a charge of $160 million to $180 million for the layoffs, mostly in the fourth quarter of fiscal 2025, it disclosed in a regulatory filing. Late in September, Washington cracked down on companies in China and other countries that use subsidiaries or other foreign affiliates to circumvent export curbs on chipmaking equipment and other goods and technology. That has made it harder for Applied Materials and its rivals to export some products and supply specific parts and services to select China-based customers without a license. The company earlier this month forecast a $600-million hit to its fiscal 2026 revenue due to the expanded curbs. It said it started informing its employees about the layoffs earlier on Thursday. It had 35,700 full-time staff as of October 27, 2024, according to its annual report. "Our goal is to continue to transform how we work, move faster, simplify decision-making, and focus on what matters most as we prepare Applied Materials for significant growth in the coming years," CEO Gary Dickerson said in a memo to employees. (Reporting by Aditya Soni in Bengaluru; 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.)

(Reuters) -Anthropic said on Thursday its Claude models would be trained using up to one million of Google's artificial intelligence chips, worth tens of billions of dollars, as it aims to improve its generative AI offerings in a fast-moving landscape. The deal underscores the immense computing demands of AI firms for training, deployment and continuous inference in GenAI. To secure infrastructure swiftly, developers have been rushing into multi-billion-dollar agreements. Alphabet's Google, also a backer of Anthropic, will provide additional cloud computing services to the startup behind the Claude chatbot. For Google, the deal comes as it is expanding external availability of its in-house tensor processing units, or TPUs, which were historically reserved for internal use. It rents the TPUs through Google Cloud. Anthropic said it chose the TPUs due to their price-performance and efficiency and its existing experience in training and serving its models with the processors. Reuters exclusively reported earlier in October that Anthropic is projecting to more than double and potentially nearly triple its annualized revenue run rate next year, fueled by the rapid adoption of its enterprise products. (Reporting by Juby Babu in Mexico City and Zaheer Kachwala in Bengaluru; Editing by Shilpi Majumdar, Anil D'Silva and 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.)

(Reuters) -Applied Materials will cut about 4% of its workforce, or roughly 1,400 jobs, to streamline operations, the chip equipment maker said on Thursday as tighter U.S. export controls on semiconductors weigh on its business. The company, one of the largest U.S. makers of semiconductor manufacturing equipment, will take a charge of $160 million to $180 million for the layoffs, mostly in the fourth quarter of fiscal 2025, it disclosed in a regulatory filing. Late in September, Washington cracked down on companies in China and other countries that use subsidiaries or other foreign affiliates to circumvent export curbs on chipmaking equipment and other goods and technology. That has made it harder for Applied Materials and its rivals to export some products and supply specific parts and services to select China-based customers without a license. The company earlier this month forecast a $600-million hit to its fiscal 2026 revenue due to the expanded curbs. It said it started informing its employees about the layoffs earlier on Thursday. It had 35,700 full-time staff as of October 27, 2024, according to its annual report. "Our goal is to continue to transform how we work, move faster, simplify decision-making, and focus on what matters most as we prepare Applied Materials for significant growth in the coming years," CEO Gary Dickerson said in a memo to employees. (Reporting by Aditya Soni in Bengaluru; 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.)