विज्ञान

By Hannah Lang and Elizabeth Howcroft NEW YORK/PARIS (Reuters) -A race by crypto companies to sell tokens pegged to stocks is raising alarm bells among traditional financial firms and regulatory experts who warn that the fast-growing novel products pose risks to investors and market stability. Buoyed by President Donald Trump's pro-crypto stance and his administration's push for friendly regulations, the crypto industry is rushing to capitalize on a global surge in enthusiasm for the sector.  Robinhood, Gemini and Kraken among others have launched tokenized stocks in Europe, while Coinbase, Robinhood and startup Dinari are seeking approval to launch similar products in the United States. Nasdaq, meanwhile, last month became the first major exchange to propose offering tokenized shares.  The industry says tokenized shares — blockchain-based instruments that track traditional equities — could revolutionize stock markets by allowing shares to be traded 24/7 and settled instantly, boosting liquidity and reducing transaction costs. The combined value of tokenized public stocks geared toward retail investors as of September grew to $412 million, compared with just a few million dollars 12 months ago, according to tokenization tracker RWA.xyz. Although many products are marketed like stocks, they rarely offer the same rights, disclosures and protections as traditional equities. Instead, they more closely resemble riskier derivatives, according to a Reuters review of several products and interviews with a dozen industry executives and legal experts. That increases the hazards for investors, while tokenization more broadly could undermine market integrity and fragment liquidity if left unsupervised, critics say. "You're buying exposures to those shares through creating some sort of synthetic instrument," said Diego Ballon Ossio, a partner at law firm Clifford Chance in London. "A lot of the burden gets shifted on you to understand what exactly it is that you're buying." A few companies have issued their own experimental stock tokens on the blockchain – software that acts as a shared digital ledger – but most tokenized shares are pegged to public companies and issued by third parties like Ondo Global Markets and Dinari. Some tokens are backed 1:1 by underlying stocks, while others provide economic exposure through derivatives.  The industry is divided over which regulations apply to stock tokens, and investor rights and protections vary. Often, the products provide no ownership, voting rights or traditional dividends, while creating counterparty risk exposure to the token issuer.  For example, there are multiple tokens pegged to Nvidia and Tesla with a range of structures and terms and conditions. "The fact that different tokenized offerings have different rights and different disclosures … that's a real big worry," said Gabriel Otte, CEO of Dinari, which offers 1:1 collateralization.  Robinhood in June launched trading in tokens pegged to public companies and said it plans to offer tokenized stocks of private companies. To promote the launch, it gave away tokens pegged to OpenAI. Those tokens are derivative contracts backed by Robinhood's ownership of fund units in a special-purpose vehicle that holds OpenAI convertible notes, according to its terms and conditions. The announcement drew pushback from OpenAI, which said it had not blessed the offering. It also prompted scrutiny from Robinhood's European regulator. Johann Kerbrat, general manager of Robinhood Crypto, said the company clearly flags that its tokens are derivatives. "It's just one step forward to be able to have the benefits of no longer having multiple days to settle," he added.  While Robinhood is issuing public company tokens on the blockchain, it is not yet settling the trades on the blockchain, a spokesperson said. Gemini declined to comment. CORE INVESTOR PROTECTIONS In Europe, Robinhood, Kraken and others operate under the "MiFID" derivatives rules but some legal experts say that law is insufficient to oversee the novel products. Trump's crypto-friendly chair of the U.S. Securities and Exchange Commission, Paul Atkins, has indicated the agency plans to grant would-be issuers exemptions from securities rules.  That plan is facing opposition from powerful Wall Street players including Citadel Securities and the Securities Industry and Financial Markets Association, which say such major structural changes should go through a formal rulemaking process.  "Just because a security is represented on blockchain, that doesn't change the core investor protections and other provisions that apply to securities," said Peter Ryan, head of international capital markets at SIFMA.  In a July letter to the SEC, Citadel Securities raised concerns that tokenization would siphon liquidity away from public markets.  Spokespeople for the SEC declined to comment, while Citadel Securities did not provide comment beyond the letter.  A spokesperson for the European Securities and Markets Authority, which helps oversee MiFID, said it was aware of the potential risks of tokenization and was monitoring developments.  The World Federation of Exchanges recently urged regulators to crack down on tokenization, citing insufficient investor protections and liquidity fragmentation, although the group told Reuters it supports Nasdaq's proposal because it would treat tokens like traditional stocks. Coinbase is also in talks with the SEC about launching tokenized securities that would similarly grant investors the full legal rights and benefits associated with conventional stocks, according to a source familiar with the matter. Other issuers said they hew closely to traditional securities, anti-money laundering, bankruptcy protections and other rules.  Mark Greenberg, Kraken's global head of consumer, said the company offered the "gold standard" including 1:1 collateralization and investor disclosures, while dismissing derivative offerings as "IOUs." "Done right, tokenization enhances investor protections, rather than eroding them," said Ian De Bode, chief strategy officer at Ondo Finance.  (Reporting by Hannah Lang in New York and Elizabeth Howcroft in Paris; Editing by Michelle Price 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) -OpenAI and Anthropic are exploring the use of investor funds to settle potential multibillion-dollar lawsuits, the Financial Times reported on Wednesday. Copyright owners have launched a series of high-stakes lawsuits against tech firms, including OpenAI, Microsoft, and Meta Platforms, alleging unauthorized use of their material to train AI systems. OpenAI, which has tapped Aon, has secured coverage of up to $300 million for emerging AI risks, the report said, citing people familiar with the company's policy. However, another person disputed this figure, claiming it was significantly lower, the newspaper said, adding that all agreed the coverage falls far short of what is needed to protect against potential losses from a series of multibillion-dollar legal claims. Kevin Kalinich, head of cyber risk at Aon, told the newspaper that the insurance sector broadly lacks "enough capacity for (model) providers." OpenAI has considered "self-insurance" by setting aside investor funding, with discussions about establishing a "captive," a ringfenced insurance vehicle used by large companies to manage emerging risks, the report said. OpenAI, Anthropic and Aon did not immediately respond to Reuters' request for comment. Reuters could not immediately verify the report. Last month, a federal judge in California preliminarily approved a landmark $1.5 billion settlement of a copyright class action brought by a group of authors against Anthropic. Anthropic is partly using its own funds to cover potential settlements, the FT report said, citing a person with knowledge of the matter. (Reporting by Bipasha Dey in Bengaluru; Editing by Mrigank Dhaniwala)

(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) -OpenAI and Anthropic are exploring the use of investor funds to settle potential multibillion-dollar lawsuits, the Financial Times reported on Wednesday, citing people familiar with the matter. Reuters could not immediately verify the report. (Reporting by Bipasha Dey in Bengaluru; Editing by Mrigank Dhaniwala)

(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 Francesco Guarascio and Phuong Nguyen HANOI (Reuters) -Honda's dominance of Vietnam's motorbike market is under threat as Hanoi pushes for a rapid switch to electric vehicles in a boost to local EV maker VinFast, according to industry data and the findings of a consumer survey. Vietnam's Prime Minister Pham Minh Chinh issued a directive in July that bans after 12 months fossil-fuel-powered motorbikes from the centre of the capital Hanoi and extends the scope of the ban from 2028. Honda and other motorbike makers have written to authorities criticising the changes as being too fast, according to two people with direct knowledge of the matter. The Japanese company's sales in Vietnam fell by nearly 22% in August from July, a 13% drop on the year, after the government's ban was announced, industry data showed. Partly in response to the government's policies, residents of Hanoi and Ho Chi Minh City, Vietnam's two largest cities, plan to choose an electric model for their next motorbike purchase, according to a survey conducted in September by market research firm Asia Plus Inc.  Representatives for Honda in Vietnam did not reply to requests for comment. The Vietnamese ministry responsible for transport did not reply to a request for comment but the government has said the new measures were needed to address air pollution in Hanoi, which at times tops the list of the world's most polluted cities.   LION'S SHARE OF THE MARKET Vietnam's two-wheeler market stands at $4.6 billion in 2025 and is projected to reach $6 billion by 2030, according to market research firm Mordor Intelligence. Petrol motorbikes dominate Vietnam's market, one of the world's largest. Honda sold 2.6 million vehicles in the country last year, more than 80% of the total. VinFast said it delivered 71,000 electric motorbikes. Honda, the world's largest manufacturer of motorbikes, is also producing electric models but it is not clear how many it sells in Vietnam. Of the 200 people polled in the survey, 54% said they would choose an electric motorbike for their next purchase, and only 24% would opt for a gasoline model. The main reason given for preferring an electric option was the government's policies to endorse EVs. In Hanoi, the share of interviewees saying they would choose an electric motorbike was 60%. Half of the poll respondents said they would consider buying a Honda motorbike and 32% said they would opt for a VinFast model. More than 80% said they already owned a Honda vehicle and 4% said they had a VinFast motorbike. "The rapid policy support, including phased bans, incentives, and public messaging, has significantly shaped consumer attitudes and could accelerate a market shift faster than traditional manufacturers anticipated," said Kengo Kurokawa, the head of Asia Plus. (Reporting by Francesco Guarascio; Additional reporting by Phuong Nguyen in Hanoi and Daniel Leussink in Tokyo; Editing by Muralikumar Anantharaman)

(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) -The Federal Bureau of Investigation's Washington field office is investigating a series of alleged infiltration of prominent U.S. law firms by Chinese hackers, the New York Times reported on Tuesday. One of those firms, Williams & Connolly, told clients that Chinese hackers had infiltrated some of its computer systems, the newspaper reported, citing two people briefed on the matter. The FBI did not immediately respond to a request for comment. (Reporting by Jasper Ward)

(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) -British police said on Tuesday that they had arrested two people on suspicion of computer misuse and blackmail following a cyberattack on a London childcare company in which data on more than 8,000 children was stolen. The hackers declined to say how much money they were demanding from Kido International, which operates 18 nurseries in Greater London that typically serve babies up to 5-year-olds. The gang, which identifies itself as Radiant, reported on the attack on its dark web portal last month. It evidenced its claim by publishing the names, photos, home addresses and family contact information of 10 children it said attended one of Kido's centres. London's Metropolitan Police said the two men, aged 17 and 22, were arrested and taken into custody, where they will remain for questioning. The arrests followed an operation at several residential properties in the town of Bishop's Stortford. The hack, which raises serious concerns about child safeguarding and data privacy, was the latest in a string of serious ransomware incidents in Britain that have rocked businesses in Britain this year. (Reporting by Catarina Demony; Editing by Cynthia Osterman)

(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 Abhirup Roy and Harshita Mary Varghese (Reuters) -Tesla rolled out "affordable" versions of its best-selling Model Y SUV and its Model 3 sedan, but the starting prices of $39,990 and $36,990 were too high, some said, to attract a new class of buyers to the electric vehicle brand. CEO Elon Musk has touted the car as a way to tap a wider swath of buyers, saying last year that a price below $30,000, after incentives, was the key. Tesla is trying to reverse falling sales of its aging lineup amid rising competition in Europe and China, and the loss of a $7,500 U.S. tax credit. The new cars drop some premium finishes and features but offer driving ranges above 300 miles. Tesla's stock closed down 4.5% and Tesla bull Dan Ives, an analyst at Wedbush, said he was disappointed that the cars were only about $5,000 cheaper than the next-level trims of the models. The new Standard versions cost more than the cheapest U.S. models in September, when the now-expired tax credit was included. Musk has for years promised mass-market vehicles, though last year he canceled plans for an all-new $25,000 EV, Reuters first reported. Instead, he chose to build lower-priced versions based on Tesla's current models, sparking concerns among investors and analysts that the cheaper cars would cannibalize sales of existing vehicles and limit growth. "It's basically a pricing lever and not much of a product catalyst," said Shay Boloor, chief market strategist at research firm Futurum Equities. "I don't see it as unlocking new demand at scale." Fearing a drop in demand after losing the tax credits, some automakers in the U.S. have already cut prices while some others have figured out mechanisms to effectively extend the benefit of the incentives. Investors and analysts said dropping the price below $40,000 will help drive sales as the latest offerings compete next year with Chevrolet's Equinox, Hyundai Ioniq 5 and Kia's EV4. Some, though, expected a bigger slash by Tesla to below $30,000. "I just don't know that this is enough," said Shawn Campbell, an advisor at Camelthorn Investments. "Longer term, this news doesn't solve the problem posed by lower cost Chinese competitors in global markets. In my opinion, Tesla needs a sub-30k EV." In Europe, where Musk's far-right political views have undermined brand loyalty, the new entries will be up against more than a dozen electric and plug-in-hybrid models with price tags below $30,000. Sales in the September-ending quarter rose to a record as consumers rushed to take advantage of the EV tax credit before its expiration on September 30, but expectations are that they will slow down for the rest of the year, unless the affordable car comes to the rescue. "The desire to buy the car is very high. (It's) just (that) people don't have enough money in the bank account to buy it," Musk said in July during Tesla's second-quarter earnings call. "So the more affordable we can make the car, the better." 'A WALMART HOODIE' FOR TESLA Both Standard versions offer 321 miles (516 km) of range and less powerful acceleration than the current higher trims, called Premium. The Model 3 includes a smaller battery pack. Both cars can be ordered immediately, with deliveries set to start between December 2025 and January 2026 for many locations, Tesla's website showed.  The Standard versions do not come with Autosteer, Tesla's driver assistance system, touchscreens and seat heating for rear passengers. Tesla has also removed the LED lightbar in the cheaper Model Y. Both come with textile seats, with vegan leather available for the Model 3, and manually adjusted side-view mirrors.  Musk has been pivoting the company toward artificial intelligence, focusing on robotaxis and humanoid robots, but the new cars are key to hopes of near-term revenue growth. "For the market, this is Tesla dropping its Steve Jobs turtleneck and slipping into a Walmart hoodie," said Michael Ashley Schulman, chief investment officer at Running Point. "It's no longer the cool rebel at the edge of innovation – it's the establishment trying to play both Tesla and Toyota at once." (Reporting by Abhirup Roy in San Francisco and Akash Sriram and Harshita Mary Varghese in Bengaluru; Editing by Peter Henderson, Richard Chang, Sriraj Kalluvila 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 Milana Vinn (Reuters) -Confluent is exploring a sale after attracting acquisition interest, according to three people familiar with the matter, the latest data infrastructure company to draw suitors for its potential in supporting artificial intelligence development. The software provider is working with an investment bank on the sale process, which is in its early stages and was instigated after both private equity firms and other technology companies expressed their interest to the company in buying it, the sources said. The company has a market capitalization of approximately $7 billion, but its stock is trading around 26% lower year-to-date. Some of the sources noted while Confluent's technology is highly sought-after, it became vulnerable to takeover approaches when its stock price dived in July, when it reported losing business from a large customer. The sources cautioned that no deal is guaranteed and Confluent could ultimately remain independent. They also spoke on condition of anonymity because the discussions are confidential. Confluent did not immediately respond to a comment request. Mountain View, California-based Confluent provides technology needed to manage massive, real-time data streams for artificial intelligence models. The company was founded by the original creators of Apache Kafka, a popular open-source technology that helps companies process huge torrents of data in real time, from bank transactions to website clicks. Confluent commercializes Kafka, providing enterprise-grade features, services, and support to major corporations worldwide. The interest in Confluent highlights a surge in demand for data infrastructure companies, fueled by the corporate race to develop generative artificial intelligence. In May, Salesforce agreed to acquire software maker Informatica for approximately $8 billion to bolster its AI capabilities by integrating Informatica's data management, integration, and governance tools into the Salesforce platform. (Reporting by Milana Vinn in New York; Editing by David French 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 Stephen Nellis SAN FRANCISCO (Reuters) -U.S. lawmakers are calling for broader bans on chipmaking equipment to China after a bipartisan investigation found that Chinese chipmakers had purchased $38 billion of sophisticated gear last year. Inconsistencies in rules issued by the United States, Japan and the Netherlands have led to non-U.S. chip equipment manufacturers selling to some Chinese firms that U.S. companies could not, according to a report published on Tuesday by the U.S. House of Representatives Select Committee on China. The committee called for broader bans by the U.S. and its allies on chipmaking tool sales to China, rather than narrower bans on sales to specific Chinese chipmakers. The $38 billion was purchased from five top semiconductor manufacturing equipment suppliers, without breaking the law, a 66% increase from 2022, when many of the tool export restrictions were introduced. That accounted for nearly 39% of the aggregate sales of Applied Materials, Lam Research, KLA, ASML and Tokyo Electron, the report found. "These are the sales that made China increasingly competitive in the manufacture of a wide range of semiconductors, with profound implications for human rights and democratic values around the world," the report said. Both U.S. Democratic and Republican administrations have sought to restrict China's ability to make microchips – crucial to fields such as artificial intelligence and military modernization. The two economic superpowers are also vying to sell advanced technology such as AI data centers to other nations. Mark Dougherty, president of Tokyo Electron's U.S. unit, said the industry's China sales have started to decline this year, in part due to new regulations and welcomed more coordination between the U.S. and Japanese governments. "I think it’s clear, from a U.S. perspective, there’s an outcome that is still desired that has not yet been achieved," Dougherty told Reuters in an interview. ASML and KLA declined to comment. Applied Materials and Lam Research did not respond to a request for comment. The committee said the toolmakers cooperated with the committee on the report and were informed of its findings. Three Chinese firms that have become major customers of toolmakers – SwaySure Technology Co, Shenzhen Pengxinxu Technology Co and SiEn (Qingdao) Integrated Circuits Co – are of particular security concern. They were flagged last year by the congressional committee's leaders, Chairman John Moolenaar, a Michigan Republican, and Ranking Member Raja Krishnamoorthi, an Illinois Democrat, in a letter to the Commerce Department alleging ties to a secret network aiding Huawei Technologies, and U.S. officials barred exports to them in December.  The report recommended that the broader bans include tighter restrictions on components that China could use to build its own chipmaking tools. "China is attempting to rewrite the entire supply chain," said Craig Singleton, a senior fellow at the Foundation for Defense of Democracies, a think tank. "What used to be niche tool segments are now battlegrounds." (Reporting by Stephen Nellis in San Francisco; Editing by Peter Henderson, Shri Navaratnam and Edwina Gibbs)

(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, the artificial intelligence startup backed by Alphabet's Google and Amazon.com, said on Tuesday it will open its first office in India next year, aiming to tap into the country's growing appetite for AI tools. The move comes as AI adoption in India accelerates, fueled by rising enterprise tech spending, a growing pool of skilled talent, and increasing investor interest. India, home to nearly a billion internet users, is becoming a major battleground for global AI players. The Asian country has emerged as Anthropic's second-largest consumer market for its chatbot Claude, which competes with OpenAI's ChatGPT and is noted for its strong coding capabilities. Anthropic currently offers both free and paid tiers of Claude in India but has not yet introduced local currency pricing. Co-founder and CEO Dario Amodei is scheduled to visit India this week to meet with public officials and corporate partners, the $183 billion company said. The new office will be located in Bengaluru, widely recognized as the technology hub of India, and operations will start early 2026, Anthropic said. The location would serve as its second office in the Asia Pacific region after Tokyo, it added. OpenAI, backed by Microsoft, was formally registered as a legal entity in India in 2025 and plans to open its first India office in New Delhi later this year. OpenAI and Anthropic face strong competition in India from rivals such as Google's Gemini and AI startup Perplexity, both of which have launched offerings that make their advanced plans free for many users in the market. Last month, Anthropic announced plans to triple its international workforce to meet a surge in demand for its Claude AI models outside the United States. (Reporting by Mrinmay Dey in Bengaluru; Editing by Alan Barona, Janane Venkatraman and Sherry Jacob-Phillips)

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