विज्ञान

WASHINGTON (Reuters) -U.S. government officials said on Wednesday that federal networks are being targeted by an unidentified "nation-state cyber threat actor" that's trying to exploit vulnerabilities in products made by the cybersecurity company F5. In a statement and an accompanying emergency directive, the Cybersecurity and Infrastructure Security Agency said hackers had compromised F5's systems and extracted files, including a portion of its source code and information about vulnerabilities, and could use the knowledge as a roadmap to break into F5 devices and software, something that could eventually lead to a full compromise of the targeted networks.  "The cyber threat actor presents an imminent threat to federal networks" using F5 products, CISA said. CISA's Executive Assistant Director for Cybersecurity Nick Andersen told reporters that government officials were being ordered to identify F5's devices on their network and apply urgent updates. Andersen encouraged others to do the same, noting that "the risk of this vulnerability extends to every organization and sector that's using this product." Andersen refused to say who the hackers were and said there had so far been no evidence of any compromise at a U.S. civilian agency.  Earlier, F5 said it had detected unauthorized access to certain company systems by a threat actor, but the breach had no impact on its operations. The company discovered the intrusion on August 9 and took "extensive actions" to contain the threat, engaging external experts, including CrowdStrike, Mandiant, NCC Group and IOActive, to assist with the investigation, it said in a filing with the U.S. Securities and Exchange Commission. The company said it found no signs that its software development process had been tampered with. F5, which has clients across the private and public sector, said information from a few customers was involved in the breach, and it was reaching out to those affected directly. The company continues to strengthen its security controls and infrastructure following the incident, it said, adding that the U.S. Department of Justice had approved a delay in publicly disclosing the breach until September 12, citing national security considerations. British authorities also issued an alert urging F5 users to update their software. (Reporting by Akash Sriram in Bengaluru and Raphael Satter in Washington; Editing by Shilpi Majumdar, David Gregorio and Diane Craft)

(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 Suzanne McGee and Ateev Bhandari (Reuters) -Volatility Shares, an issuer of exchange-traded funds, filed on Wednesday to launch a total of 27 highly leveraged ETFs, including the first-ever proposed 5x ETF for the U.S. market, at a time of rising caution over inflated asset prices as markets continue their upward swing. A 5x target means that an ETF would seek to quintuple the daily return of an underlying single stock. The single stocks involved include tech heavyweights Tesla, Nvidia, AMD, Amazon and Palantir, representing some of the greatest beneficiaries of the AI-spending fueled market rallies. The list also includes Coinbase and crypto treasury pioneer Strategy, two key players from a sector that has produced stellar wins for the IPO market this year. Volatility Shares has filed for a 3x and a 5x offering on Strategy, where existing issuers have struggled to manage a 2x product. Until now, the U.S. Securities and Exchange Commission has approved single-stock leveraged ETFs with a maximum of 2x. "You have to designate a reference benchmark and use that to calculate 'value at risk,' or the chance that the product will be unduly risky for either investors or the market," said Rahul Sharma, president of financial data provider Indxx, referring to the SEC's rules about launching such products. Sharma noted that recently the SEC changed guidance on what assets could be used as a reference benchmark to make it easier to win approval, and that the pace of new leveraged ETF filings has been growing. According to StockTwits, there are now some 900 leveraged products, accounting for 33% of all new ETFs but only 1% of the U.S. ETF industry's $12 trillion in assets. However, firms have "enormous latitude in defining these calculations," said ETF.com President David Nadig. Volatility Shares declined to comment on its plans to calculate "value at risk." According to Nadig, a partisan gridlock in Washington has also provided some cover for the filing and potential launching of these ETFs. "If the SEC came back and sat at their desks, they'd put the kibosh on this. In the normal course of business you'd expect these to be shot down quickly," Nadig added. A Reuters analysis of SEC filings found that Volatility has proposed its filings to go effective 75 days after submitting. Such time-based flipping of filing status has previously been used by companies looking to IPO during a shutdown as well. According to Indxx's Sharma, the former rules "prevented the launch of some of the highest volatility products," while the new guidance "introduces more systemic risk." Concerns are piling on that a correction is on the horizon, and products designed to amplify upside could end up achieving the same for a crash. The crypto market experienced the largest liquidations in history, with more than $19 billion wiped out across leveraged positions late on Friday after U.S. President Donald Trump escalated his trade rhetoric against China. Tech stocks are also not insulated from such shocks and a highly leveraged market positions can amplify the selloff, as seen on Friday, when some leveraged investors were forced to close their positions after their collateral fell below certain thresholds. A report from JPMorgan estimated that some $26 billion of selling from leveraged ETFs at Friday's close drove the overall market even lower. "These products are a terrible way to get this kind of leverage," Nadig said. (Reporting by Ateev Bhandari in Bengaluru and Suzanne McGee; 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 Krystal Hu and Deepa Seetharaman (Reuters) -Artificial intelligence startup 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, according to two people familiar with the matter. The company is on track to meet an internal goal of $9 billion in annual revenue run rate – a calculation of annual revenue extrapolated from the current sales pace – by the end of 2025, the people said. For 2026, Anthropic has set even more aggressive targets: a base case of more than doubling to $20 billion in annualized revenue and a best case of as much as $26 billion, the people said, requesting anonymity to discuss private figures. Anthropic told Reuters its annual revenue run rate is approaching $7 billion this month, but declined to comment on future projections. The company has said its annual revenue run rate was more than $5 billion in August. Anthropic debuted a new version of its cheapest AI model, Haiku, on Wednesday, as part of a broader effort to appeal to companies that are looking for capable AI systems that are dramatically cheaper than its more advanced models. The Haiku 4.5 model sells for about one-third the price of Sonnet 4, one of its medium-sized models. AI DEMAND SOARS The revenue projections underscore continued strong demand for generative AI tools among businesses and help explain investor enthusiasm, even as AI spending, especially in infrastructure buildout, comes under scrutiny. Some people worry the level of investment might be unsustainable. Fueling the expansion is the uptake of enterprise products, which are built for organizations. Anthropic has more than 300,000 business and enterprise customers, which account for about 80% of its revenue. The company sells access to its models via application programming interfaces, which are tools that help different software communicate. It offers products including the code‑generation tool Claude Code, which has reached an annualized revenue run rate of nearly $1 billion since its launch earlier this year, one of the people added. The revenue trajectory positions Anthropic as a rival to ChatGPT maker OpenAI. OpenAI said it crossed $13 billion in annualized revenue in August, and is on pace to achieve over $20 billion by end of the year, driven by the continued growth of ChatGPT, the generative AI assistant that has more than 800 million weekly active users. Anthropic's revenue milestones follow a period of brisk fundraising and valuation gains for the San Francisco-based startup. The company was recently valued at $183 billion after raising $13 billion in a Series F round led by ICONIQ, more than doubling its $61.5 billion valuation in March. Backed by technology companies including Alphabet's Google and Amazon.com, Anthropic develops the Claude family of large language models, which compete with OpenAI's GPT series. The company emphasizes AI safety and building models for enterprise use cases. Its models have helped power a boom in code‑generation startups such as Cursor. Anthropic has been expanding sales to governments and growing outside the U.S. In August, it said it would offer its Claude model to the U.S. government for $1. The company plans to open its first office in Bengaluru, India, in 2026 — its second-largest market after the U.S.— and intends to triple its international workforce and expand its applied AI team fivefold this year to meet rising demand. (Reporting by Krystal Hu and Deepa Seetharaman in San Francisco; editing by Kenneth Li and 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.)

WASHINGTON (Reuters) -U.S. government officials said on Wednesday that federal networks are being targeted by an unidentified "nation-state cyber threat actor" that's trying to exploit vulnerabilities in products made by the cybersecurity company F5. In a statement and an accompanying emergency directive, the Cybersecurity and Infrastructure Security Agency said hackers had compromised F5's systems and extracted files, including a portion of its source code and information about vulnerabilities, and could use the knowledge as a roadmap to break into F5 devices and software, something that could eventually lead to a full compromise of the targeted networks.  "The cyber threat actor presents an imminent threat to federal networks" using F5 products, CISA said. CISA's Executive Assistant Director for Cybersecurity Nick Andersen told reporters that government officials were being ordered to identify F5's devices on their network and apply urgent updates. Andersen encouraged others to do the same, noting that "the risk of this vulnerability extends to every organization and sector that's using this product." Andersen refused to say who the hackers were and said there had so far been no evidence of any compromise at a U.S. civilian agency.  Earlier, F5 said it had detected unauthorized access to certain company systems by a threat actor, but the breach had no impact on its operations. The company discovered the intrusion on August 9 and took "extensive actions" to contain the threat, engaging external experts, including CrowdStrike, Mandiant, NCC Group and IOActive, to assist with the investigation, it said in a filing with the U.S. Securities and Exchange Commission. The company said it found no signs that its software development process had been tampered with. F5, which has clients across the private and public sector, said information from a few customers was involved in the breach, and it was reaching out to those affected directly. The company continues to strengthen its security controls and infrastructure following the incident, it said, adding that the U.S. Department of Justice had approved a delay in publicly disclosing the breach until September 12, citing national security considerations. British authorities also issued an alert urging F5 users to update their software. (Reporting by Akash Sriram in Bengaluru and Raphael Satter in Washington; Editing by Shilpi Majumdar, David Gregorio and Diane Craft)

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

WASHINGTON (Reuters) -U.S. government officials said on Wednesday that federal networks are being targeted by an unidentified "nation-state cyber threat actor" that's trying to exploit vulnerabilities in products made by the cybersecurity company F5. In a statement and an accompanying emergency directive, the Cybersecurity and Infrastructure Security Agency said hackers had compromised F5's systems and extracted files, including a portion of its source code and information about vulnerabilities, and could use the knowledge as a roadmap to break into F5 devices and software, something that could eventually lead to a full compromise of the targeted networks.  "The cyber threat actor presents an imminent threat to federal networks" using F5 products, CISA said. CISA's Executive Assistant Director for Cybersecurity Nick Andersen told reporters that government officials were being ordered to identify F5's devices on their network and apply urgent updates. Andersen encouraged others to do the same, noting that "the risk of this vulnerability extends to every organization and sector that's using this product." Andersen refused to say who the hackers were and said there had so far been no evidence of any compromise at a U.S. civilian agency.  Earlier, F5 said it had detected unauthorized access to certain company systems by a threat actor, but the breach had no impact on its operations. The company discovered the intrusion on August 9 and took "extensive actions" to contain the threat, engaging external experts, including CrowdStrike, Mandiant, NCC Group and IOActive, to assist with the investigation, it said in a filing with the U.S. Securities and Exchange Commission. The company said it found no signs that its software development process had been tampered with. F5, which has clients across the private and public sector, said information from a few customers was involved in the breach, and it was reaching out to those affected directly. The company continues to strengthen its security controls and infrastructure following the incident, it said, adding that the U.S. Department of Justice had approved a delay in publicly disclosing the breach until September 12, citing national security considerations. British authorities also issued an alert urging F5 users to update their software. (Reporting by Akash Sriram in Bengaluru and Raphael Satter in Washington; Editing by Shilpi Majumdar, David Gregorio and Diane Craft)

(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 Echo Wang NEW YORK (Reuters) -Meta Platforms said on Wednesday it would invest $1.5 billion in a data center in Texas, breaking ground at its 29th such facility globally, as the social media giant expands infrastructure to support artificial intelligence workloads. The data center in El Paso, Meta's third in Texas, is expected to come online in 2028, and can scale to a 1-gigawatt site – enough energy to power a city the size of San Francisco for a day – making it one of the largest planned data center campuses in the U.S. Large-scale cloud service providers, called hyperscalers, have been racing to build AI infrastructure, with Amazon, Alphabet, Meta, 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 new facility is projected to create about 100 jobs once operational, with over 1,800 construction workers expected onsite at peak construction, Meta said in a statement. The company cited El Paso's robust electrical grid and skilled workforce as factors in the site selection. Meta has invested over $10 billion in Texas and employs more than 2,500 people across the state, the company said. These figures include the latest investment.  The company is putting up $1.5 billion of its own money to fund the current phase of the El Paso site. Wednesday's announcement follows Meta's recent $29 billion off-balance-sheet deal with Pimco and Blue Owl to fund a data center campus in Louisiana. "The fastest gazelle finds their place, others follow, and Meta is the fastest gazelle in the industry, as far as I'm concerned," Jon Barela, CEO of the Borderplex Alliance, a local economic development and policy advocacy group involved in facilitating the project, told Reuters. "We've had others look at the region, other data centers, and we expect there probably to be others that will want to follow."  Barela said the Meta project first came from a referral from Texas Governor Greg Abbott’s office about four years ago, and El Paso offered a package of tax incentives and other measures to attract the company. Meta said the data center will be matched with 100% renewable energy. The facility will use a closed-loop, liquid-cooled system that continuously recycles water. Meta pledged to restore twice the water consumed by the data center to local watersheds, exceeding its 2030 goal to be "water-positive" by restoring more water than it consumes. (Reporting by Echo Wang in New York; Editing by 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.)

MILAN (Reuters) -Vantage Data Centers will invest more than 350 million euros ($407 million) to build a second campus in Italy, the U.S.-based digital infrastructure provider said on Wednesday. WHY IT'S IMPORTANT The campus aims to support the growing demand for high-performance digital infrastructure in the country and its location close to Italy's financial capital Milan will allow connectivity with other key European digital hubs, including Zurich, Frankfurt and Munich, Vantage said in a statement. BY THE NUMBERS The new campus, which will add to a first one in Milan, will consist of two data centers with a combined 32 megawatts of IT capacity. The first phase is expected to be operational in the spring of 2026. CONTEXT The Italian government has recently recognised Vantage's multiple campus developments as nationally strategic infrastructure investments. Through this program, Vantage will invest approximately 4 billion euros and employ an estimated 4,000 workers over eight years during construction, while also creating approximately 250 permanent jobs. ($1 = 0.8594 euros) (Writing by Francesca Landini; editing by Kirsten Donovan)

(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 Phoebe Seers LONDON (Reuters) -The Bank of England will only remove proposed caps on the quantity of stablecoins individuals and businesses can hold when it is confident they do not pose a threat to financial stability. The central bank has taken a tougher stance than many authorities on embracing stablecoins, a type of cryptoasset pegged to fiat currency that has soared in popularity. The crypto industry has opposed the planned caps, but BoE Deputy Governor Sarah Breeden reiterated on Wednesday they were needed because of the potential impact of significant and rapid outflows of bank deposits into stablecoins. Such outflows, she said in a speech, could lead to a "precipitous drop in credit for businesses and households if the banking system were unable to increase, at scale and at pace, its use of wholesale financing from non-banks". The BoE has previously indicated it could apply limits of 10,000 pounds to 20,000 pounds ($26,778) for individuals, and higher levels for businesses, marking it out from other major countries which do not have limits.  It is due to publish a consultation setting out updated proposals for a future regulatory regime next month. This will seek views on proposed levels for stablecoin limits.  Breeden said the BoE would expect to remove the limits "once we see that the transition no longer threatens the provision of finance to the real economy". The largest businesses would likely be exempted from the cap so that they could hold more if required, Breeden said. Under Britain's proposed regulatory framework, the BoE would only regulate systemic sterling stablecoins, or those deemed capable of becoming widely used for payments, while the Financial Conduct Authority, which has proposed a less stringent regime, would regulate the remainder.  The BoE is also working with the Treasury to introduce a resolution regime for stablecoins, so that where an issuer fails, there is continuity of services for coinholders.  Breeden rejected suggestions that Britain has been slow to regulate cryptocurrencies, saying it expects to finalise its regulatory framework next year, in line with U.S. targets. ($1 = 0.7469 pounds)  (Reporting by Phoebe Seers; Editing by Tommy Reggiori Wilkes 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.)

By Foo Yun Chee BRUSSELS (Reuters) -U.S. tech giant Microsoft looks set to avoid a French antitrust investigation into its search operations because regulators are planning to dismiss a complaint by Qwant, the French rival search engine said on Wednesday. Qwant, which has historically relied on Microsoft's Bing platform to deliver search and news results, took a grievance against Microsoft's search practices to the French regulator earlier this year. It said on Wednesday it may challenge the decision to reject its case in court or take it to other authorities. Qwant had asked the watchdog to take interim action against Microsoft while it decided if there was sufficient grounds for an investigation. Qwant Chief Executive Olivier Abecassis said the French authority's investigators at a June closed-door hearing issued a recommendation to reject the company's request for an injunction and complaint. "In that case we will be left with no other option but to pursue all available legal avenues to protect our business from Microsoft's egregious abuse, to defend our rights, and to seek compensation," he said. The French regulator declined to comment. It is expected to issue its decision in the next two weeks, though the timing may still change, people with direct knowledge of the matter said. Microsoft said the Qwant complaint lacks merit. "The complaint alleges harm to competition in the market for search, which is dominated by Google," a Microsoft spokesperson said. Qwant alleged Microsoft imposes exclusivity restrictions on Qwant in search results and search advertising, hampering the latter's ability to develop its own search engine, and also alleged it favours itself in allocating search advertising, according to the French company. Microsoft is a major player in the search-engine syndication sector, where it provides search results to smaller European rivals such as Qwant, Ecosia, DuckDuckGo and Lilo. (Reporting by Foo Yun Chee;Editing by Elaine Hardcastle)

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

WASHINGTON (Reuters) -U.S. government officials said on Wednesday that federal networks are being targeted by an unidentified "nation-state cyber threat actor" that's trying to exploit vulnerabilities in products made by the cybersecurity company F5. In a statement, the Cybersecurity and Infrastructure Security Agency said exploitation of the devices "could allow the threat actor to move laterally within an organization’s network, exfiltrate sensitive data, and establish persistent system access, potentially leading to a full compromise of targeted information systems." Earlier, F5 said it had detected unauthorized access to certain company systems by a threat actor, but the breach had no impact on its operations. The company discovered the intrusion on August 9 and took "extensive actions" to contain the threat, engaging external experts, including CrowdStrike, Mandiant, NCC Group and IOActive, to assist with the investigation, it said in a filing with the U.S. Securities and Exchange Commission. F5, a provider of cybersecurity and multi-cloud application services, said the attacker had long-term access to its internal systems used to develop BIG-IP software and stole files containing parts of the program's code and details about security flaws that had not yet been made public. The company, however, said it found no signs that key security flaws were used in attacks or that its software development process had been tampered with. F5 said information from a few customers was involved in the breach, and it was reaching out to those affected directly. The company continues to strengthen its security controls and infrastructure following the incident, it said, adding that the U.S. Department of Justice had approved a delay in publicly disclosing the breach until September 12, citing national security considerations. (Reporting by Akash Sriram in Bengaluru and Raphael Satter in Washington; Editing by Shilpi Majumdar and David Gregorio)

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