विज्ञान

(Reuters) -Microsoft on Thursday introduced new features in its digital assistant Copilot such as collaboration and deeper integration with other applications like Outlook and Google, beefing up its AI services to stave off competition. Other artificial intelligence service providers such as Anthropic and OpenAI are upgrading their models and launching products aimed at capturing a wider share of the booming AI market. If it gets permission, Copilot can see and use reasoning over a user's tabs on Microsoft's Edge browser, summarize and compare information and take action like booking a hotel. Previous browsers can also be turned into "storylines" so people can revisit older ideas. Along with the software features, the company introduced an avatar called "Mico" – a nod to Microsoft Copilot – which can show expressions and change color to make conversations feel natural, the company said. The upgrades are an attempt to boost the appeal of Microsoft's browser to gain ground on Alphabet's Google Chrome, while OpenAI also entered the browser market with Atlas. Groups turns Copilot into a shared space, able to support up to 32 people, allowing users to collaborate on writing and other projects. The company has also improved Copilot's health-related questions, addressing how to ground responses from credible sources, as concerns over misinformation from AI-generated responses rise. All the updates are live in the United States, Microsoft said, adding that it will roll them out across the UK, Canada and beyond in the next few weeks. (Reporting by Zaheer Kachwala in Bengaluru; Editing by Leroy Leo)

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

By Sara Merken (Reuters) -Two federal judges admitted in response to an inquiry by U.S. Senate Judiciary Committee Chairman Chuck Grassley that members of their staff used artificial intelligence to help prepare recent court orders that Grassley called "error-ridden." In letters released by Grassley's office on Thursday, U.S. District Judge Henry Wingate in Mississippi and U.S. District Judge Julien Xavier Neals in New Jersey said the decisions in the unrelated cases did not go through their chambers' typical review processes before they were issued. Both judges said they have since adopted measures to improve how rulings are reviewed. Neals, based in Newark, in his letter said a draft decision in a securities lawsuit "was released in error – human error – and withdrawn as soon as it was brought to the attention of my chambers." He said a law school intern used OpenAI's ChatGPT for research without authorization or disclosure. Neals said his chambers has since created a written AI policy and enhanced its review process. Reuters previously reported that research produced using AI was included in the decision, citing a person familiar with the circumstances. Wingate said in his letter that a law clerk in his court in Jackson used Perplexity "as a foundational drafting assistant to synthesize publicly available information on the docket." He said posting the draft decision "was a lapse in human oversight." Wingate had removed and replaced the original order in the civil rights lawsuit and previously declined to give an explanation, saying it contained "clerical errors." The judges did not immediately respond to requests for comment sent to their court staff. Grassley had asked the judges to explain whether AI was used in the decisions after lawyers in the cases said they contained factual inaccuracies and other serious errors. Grassley in a statement on Thursday said he commended the judges for acknowledging the mistakes and urged the judiciary to adopt stronger AI guidelines. "Each federal judge, and the judiciary as an institution, has an obligation to ensure the use of generative AI does not violate litigants' rights or prevent fair treatment under the law," Grassley said. Lawyers have also increasingly faced scrutiny from judges across the country for apparent misuse of AI. Judges have levied fines or other sanctions in dozens of cases over the past few years after lawyers failed to vet the output the technology generated. (Reporting by Sara Merken; Editing by David Bario and Marguerita Choy)

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

LONDON (Reuters) -Apple on Thursday lost a London lawsuit accusing the U.S. tech company of abusing its dominant position by charging app developers an unfair 30% commission through its app store. The Competition Appeal Tribunal (CAT) ruled against Apple after a trial of the lawsuit, brought on behalf of around 20 million iPhone and iPad users in the United Kingdom and valued at up to 1.5 billion pounds ($2.01 billion), earlier this year. Rachael Kent, the British academic who brought the case, argued Apple had made "exorbitant profits" by excluding all competition for the distribution of apps and in-app purchases. The CAT ruled that Apple had abused its dominant position by shutting out competition in the app distribution market and by "charging excessive and unfair prices in the form of the commission which it charges developers". The tribunal said members of the claimant class were entitled to damages, with how damages are to be calculated to be argued at a hearing next month. Apple – which has faced mounting pressure from regulators in the U.S. and Europe over the fees it charges developers – said it would appeal against the ruling, which it said "takes a flawed view of the thriving and competitive app economy". "This ruling overlooks how the App Store helps developers succeed and gives consumers a safe, trusted place to discover apps and securely make payments," an Apple spokesperson said. The case was the first mass lawsuit against a tech giant to come to trial under Britain's fledgling class action-style regime, with many other cases waiting in the wings. ($1 = 0.7451 pounds) (Reporting by Sam Tobin; editing by William James)

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

LONDON (Reuters) -Apple on Thursday lost a London lawsuit accusing the U.S. tech company of abusing its dominant position by charging app developers an unfair 30% commission through its App Store. The Competition Appeal Tribunal ruled against Apple after a trial of the lawsuit, brought on behalf of around 20 million iPhone and iPad users in the United Kingdom and valued at up to 1.5 billion pounds ($2.01 billion), earlier this year. The case was the first mass lawsuit against a tech giant to come to trial under Britain's fledgling class action-style regime, with many other cases waiting in the wings. ($1 = 0.7451 pounds) (Reporting by Sam Tobin; editing by William James)

(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) -Mobileye Global beat Wall Street expectations for third-quarter revenue on Thursday, as automakers ramped up orders of the company's driver-assisted chips in a rush to adopt autonomous software. The company is seeing a surge in demand for its self-driving systems after a prolonged slump as its customers clear inventory. Mobileye's customers had stocked up to avoid a shortage during the pandemic. "Continued auto production stability gives us confidence to again raise our full-year outlook, removing conservatism we had embedded earlier due to macro uncertainty," said Mobileye CEO Amnon Shashua. The company also added Volvo as a new customer, executives said on a post-earnings conference call. Still, even as demand grows, economic uncertainty lingers as the imposition of tariffs on imports of automobiles and parts has pressured many of Mobileye's customers, forcing them to readjust supply chains. Mobileye previously said it would not be directly impacted by the duties. Company executives cautioned about margin pressure next year as Mobileye ramps up volume shipments of lower-margin products, but said they would improve in 2027 as product mix shifts to higher margin chips. The chipmaker, of which Intel is the largest shareholder, has previously flagged revenue potential from the robotaxi industry as several firms eye a larger portion of the fledgling market. Mobileye has been collaborating with firms such as Lyft to roll out autonomous taxis – which analysts say will be increasingly competitive in a few years with a large market. The company reported third-quarter revenue of $504 million, beating estimates of $480.9 million, according to data compiled by LSEG. The company raised the lower end of its annual revenue forecast. It now expects annual revenue of between $1.85 billion and $1.89 billion, compared with its prior projection of between $1.77 billion and $1.89 billion. (Reporting by Zaheer Kachwala in Bengaluru; Editing by Leroy Leo)

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

By Paul Sandle LONDON (Reuters) -British ad group WPP said on Thursday it was letting brands access its AI-powered marketing platform to plan, create and publish their own campaigns. WPP Open Pro, a new edition of its AI platform WPP Open, has been designed for smaller brands that do not use full-service agencies, it said, as well as to provide new capabilities to existing clients. Advertising groups are facing the challenge of brands starting to use AI to do more of their own creative and planning work, reducing the fees they pay to agencies. WPP has appointed Cindy Rose to turn it around after it lost its crown as the world's biggest ad group to French rival Publicis last year and its shares more than halved this year. The company, which owns Ogilvy, AKQA and WPP Media, said it was the first of its peers to provide a range of integrated, AI-powered delivery models, including directly to clients. Rose, who took over from Mark Read last month, said WPP's creative and strategic talent would always be at the heart of what it did for the world's biggest brands, such as Coca-Cola. But technology was fundamentally reshaping the industry, she said, and the new platform would put its AI capabilities into the hands of a much wider array of brands and businesses. "This is about transforming how marketing is delivered, expanding our total addressable market, and giving more brands the tools they need to lead in the AI era," she said. The platform will enable brands to create channel-specific ads in seconds, it said. For example, a coffee shop chain could combine brand-approved imagery with content generated by AI to create a campaign for a new store opening. The ad could then be dropped into WPP's Open Media Studio to be published or it could be published directly to major platforms, WPP said. (Reporting by Paul Sandle; editing by William James)

(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) -Roper Technologies cut its annual profit forecast on Thursday, as it expects higher costs related to acquisitions in the third quarter, sending the software company's shares down 7%. The company has grown largely through acquisitions, establishing itself as a provider of software and automated solutions to a variety of sectors, including healthcare, transportation and education. Roper has more than $5 billion in capital available for merger and acquisitions over the next 12 months, it said. It now expects adjusted earnings per share between $19.90 and $19.95 for the year, compared with its earlier expectation of $19.90 to $20.05. The forecast assumed about 10 cents of adjusted EPS dilution from quarterly acquisitions. The company said it deployed $1.3 billion for acquisitions in the third quarter. "Our underlying earnings are trending toward the higher end of our previous guidance range despite the government shutdown and some timing delays at Neptune," CEO Neil Hunn said in a statement. Roper's water-metering unit, Neptune, is facing delays in some water utility deployments. The new copper tariff that took effect on August 1 also caused some short-term disruption at Neptune. The company's Deltek government contracting business saw a slowdown in September as agencies halted activity before the shutdown. It forecast fourth-quarter adjusted profit to be between $5.11 and $5.16 per share, below analysts' average estimate of $5.25, according to data compiled by LSEG. This forecast assumes a 5-cent impact from small acquisitions. Revenue for the third quarter ended September 30 rose 14% to $2.02 billion, which was in-line with analysts estimates. It reported adjusted EPS of $5.14, beating estimates of $5.11. Roper also unveiled a new $3 billion share repurchase program on Thursday. (Reporting by Jaspreet Singh in Bengaluru; Editing by Shinjini Ganguli and Leroy Leo)

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

By Mike Scarcella (Reuters) -U.S. Google users who won a $425 million jury verdict in a consumer privacy class action last month have asked a federal judge to force the Alphabet unit to forfeit an additional $2.36 billion in profits. The consumers in a Wednesday court filing called the amount a "conservative approximation" of Google’s allegedly ill-gotten gains after the jury found the company secretly collected app activity data from millions of users who had disabled an account tracking feature. “The jury found that Google’s conduct was highly offensive, harmful, and without consent,” the consumers told Chief U.S. District Judge Richard Seeborg in San Francisco, who must decide if disgorgement of profits is allowed and necessary in the case.  GOOGLE HAS DENIED WRONGDOING, SAID IT WILL APPEAL The plaintiffs called the $425 million damages verdict "clearly insufficient to remedy the ongoing and irreparable harm that Google’s conduct continues to inflict." Google and attorneys for the plaintiffs did not immediately respond to requests for comment. Google has denied wrongdoing and said it will appeal. It argued the data it collected was anonymized and that its privacy tools give users control over their data. The 2020 lawsuit alleged Google over an eight-year period accessed users' mobile devices to collect, save and use their data, violating privacy assurances under an account setting called Web & App Activity. The jury found Google liable on two of three privacy claims brought by the plaintiffs, who had sought more than $31 billion in damages at trial. Despite the verdict, Google has not changed its privacy disclosures or data collection practices, the plaintiffs said. Google on Wednesday asked Seeborg to decertify the class of 98 million users and 174 million devices, arguing that the claims depend on individualized factors such as app usage and user expectations. It also urged the judge to vacate the verdict, citing a lack of common issues. (Reporting by Mike Scarcella. Editing by David Bario and Mark Potter)

(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 Laila Kearney NEW YORK (Reuters) -Google has entered into the first corporate agreement to buy electricity from a U.S. power plant using carbon capture and storage in a deal to help fuel its data centers in the country's Midwest region, the technology company said on Thursday. Big Tech's plans to expand technologies like generative artificial intelligence, which will require vast amounts of electricity, have butted up against the realities of a U.S. power grid running short on supplies. That has led to a flurry of announcements in recent months by companies like Google to help fund the development of new and expanded power plants across the country.  Google's recent deals have included purchasing electricity from advanced nuclear reactors, geothermal and hydropower. It is also working with the biggest U.S. power grid, PJM Interconnection, which covers the world's largest concentration of data centers, to speed up the connection of new power supplies. Google's latest power offtake agreement involves a 400-megawatt power plant in Decatur, Illinois, which will be developed by privately held Low Carbon Infrastructure. It is expected to produce power using carbon capture, which involves trapping about 90% of CO2 emissions and injecting them underground, in the early 2030s.  Google did not disclose the financial terms of the deal, while Low Carbon Infrastructure said project financing was expected to reach final investment decision in the first half of 2026. Natural gas-fired power that is produced while capturing and storing carbon dioxide had been missing from the equation, said Michael Terrell, head of Advanced Energy at Google. "We've been really focused on advancing all these new technologies for around-the-clock clean technologies and this is an important piece of the puzzle," Terrell said. "It's a very important technology that the world needs." The Broadwing project will be built at an existing industrial site operated by agribusiness company Archer Daniels Midland, which has a history of injecting carbon dioxide underground from ethanol production.  “Broadwing demonstrates that carbon capture can be commercially viable today,” said Jonathan Wiens, CEO of Low Carbon Infrastructure. Construction is expected to last four years and create 650 union labor jobs and 100 construction management and support staff in that time, the companies said. Carbon sequestered at the site will be stored permanently in special wells that are 5,000–7,000 feet (1,524-2,133 m) underground. ADM will also have the ability to purchase power from the operation, which will initially deliver electricity to the Midcontinent Independent System Operator, which covers 15 Midwest states and multiple Google data centers. Google and Low Carbon Infrastructure said they plan to pursue additional CCS facilities in the U.S., though they did not disclose specific locations or timelines. Carbon capture and storage has been promoted by the International Energy Agency and the Intergovernmental Panel on Climate Change as a tool to reduce emissions from fossil fuel-based power and heavy industry. However, critics have questioned its cost, scalability and long-term effectiveness. (Reporting by Laila KearneyEditing by Marguerita Choy)

(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 Paul Sandle LONDON (Reuters) -British ad group WPP said on Thursday it was letting brands access its AI-powered marketing platform to plan, create and publish their own campaigns. WPP Open Pro, a new edition of its AI platform WPP Open, has been designed for smaller brands that do not use full-service agencies, it said, as well as to provide new capabilities to existing clients. Advertising groups are facing the challenge of brands starting to use AI to do more of their own creative and planning work, reducing the fees they pay to agencies. WPP has appointed Cindy Rose to turn it around after it lost its crown as the world's biggest ad group to French rival Publicis last year and its shares more than halved this year. The company, which owns Ogilvy, AKQA and WPP Media, said it was the first of its peers to provide a range of integrated, AI-powered delivery models, including directly to clients. Rose, who took over from Mark Read last month, said WPP's creative and strategic talent would always be at the heart of what it did for the world's biggest brands, such as Coca-Cola. But technology was fundamentally reshaping the industry, she said, and the new platform would put its AI capabilities into the hands of a much wider array of brands and businesses. "This is about transforming how marketing is delivered, expanding our total addressable market, and giving more brands the tools they need to lead in the AI era," she said. The platform will enable brands to create channel-specific ads in seconds, it said. For example, a coffee shop chain could combine brand-approved imagery with content generated by AI to create a campaign for a new store opening. The ad could then be dropped into WPP's Open Media Studio to be published or it could be published directly to major platforms, WPP said. (Reporting by Paul Sandle; editing by William James)

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