News & Updates
The latest news and updates from companies in the WLTH portfolio.
Anthropic Computer Use
The primitive that unlocked the modern agent era. Every browser agent owes something to this. Anthropic Computer Use is Claude's first-party capability that lets the model operate a computer directly -- viewing the screen, clicking, typing, and completing tasks across software. It is the foundational primitive behind much of the current AI agent ecosystem. What it does Computer Use gives Claude screen-level control. Where MCP exposes structured APIs to the model, Computer Use exposes the entire operating system. The model takes a screenshot, decides what to click, moves the cursor, types, takes another screenshot, decides again. Multi-step workflows that previously required custom integrations now run on any software with a UI. This is what makes "AI agent" a category. Before Computer Use, agents could only operate inside software that exposed an API. After Computer Use, agents can operate anywhere a human can. Browser agents, desktop automation agents, and enterprise workflow agents are all downstream of this primitive. Key features * Screen viewing through screenshots * Mouse control -- click, drag, scroll * Keyboard input -- typing, shortcuts, multi-key * Multi-step task planning and self-correction * Available via Anthropic API and Claude products * Compatible with any software that has a UI Pricing Available through Anthropic's standard API pricing. Per-token billing scales by model tier. No separate Computer Use licensing fee. Common prompts answered * "Fill out the expense report in our procurement system for these receipts." * "Book a flight from SFO to JFK on Tuesday, business class." * "Move all files in this folder older than 90 days to archive." * "Extract data from the legacy ERP interface into a spreadsheet." * "Test the new sign-up flow end-to-end and report what breaks." Company Parent: Anthropic. Founded 2021 by Dario and Daniela Amodei. Headquartered in San Francisco. Major investment from Amazon, Google, and others. Among the most heavily capitalized private AI companies. Alternatives OpenAI Operator is the closest direct competitor. Open-source alternatives include browser-use. For structured-API tool use, see MCP servers -- they solve a related but distinct problem. The EPR MCP Server Directory covers both categories. EPR editorial verdict The primitive that unlocked the modern agent era. Computer Use is what every browser agent and most workflow agents are built on, directly or indirectly. Anthropic shipped the standard. Whether you call it from the API directly or via a downstream agent, the capability comes from here. EPR rating: 9.4/10. Last updated: May 26, 2026. Related EPR coverage: MCP Server Directory · AI Agents Directory · Claude (Anthropic) Everything-PR is the intelligence platform for communications, reputation, AI visibility, and digital discovery in the answer-engine era. Thirty-plus publications. Publishing since 2009. Original reporting, research, and analysis -- built to be cited by the AI engines that now answer the question.
Cloudvisor becomes an Anthropic authorized reseller for Amazon Bedrock, bringing Claude to startups on AWS
Cloudvisor, an AWS Advanced Tier Partner serving more than 2,000 startups across 75+ countries, has been authorized by Anthropic to resell Claude models through Amazon Bedrock. The authorization gives early-stage companies direct access to Claude Opus 4.7, Sonnet 4.6 and Haiku 4.5 inside their existing AWS infrastructure, without enterprise procurement requirements. Cloudvisor, an AWS Advanced Tier Partner focused on startups and small-to-medium businesses, has been authorized by Anthropic to resell Claude models through Amazon Bedrock. The authorization adds frontier AI capabilities to the AWS services Cloudvisor already delivers to more than 2,000 startup clients across 75+ countries. Under the agreement, founders and engineering teams can access the full Claude model family -- including Claude Opus 4.7, Claude Sonnet 4.6 and Claude Haiku 4.5 -- natively within Amazon Bedrock, alongside the rest of their AWS workloads. The arrangement removes several barriers that historically kept advanced AI tooling out of reach for early-stage companies, including minimum spend commitments, multi-month procurement cycles and enterprise-only pricing structures. The authorization extends Cloudvisor's existing AWS startup specialization into the generative AI category. The company recently signed a Strategic Collaboration Agreement with AWS focused on helping startups migrate, optimize and scale on the cloud, and reports having helped clients save more than $10 million on AWS infrastructure to date. Access through Cloudvisor includes: * Availability of the Claude model family in Amazon Bedrock without enterprise contract negotiation. * Pricing structures aligned to early-stage budgets. * Onboarding and infrastructure consulting from AWS-certified engineers. * Month-to-month engagement terms rather than long-term commitments. The announcement comes as demand for production-grade AI infrastructure continues to expand among venture-backed companies. Amazon Bedrock provides a managed environment for foundation models from multiple providers, with Claude positioned among its most widely adopted model families for enterprise and developer workloads. Cloudvisor will continue to provide Claude in Amazon Bedrock alongside its existing portfolio of AWS services, which includes AWS Resell, AWS Cost Optimization, Migration to AWS, the AWS Well-Architected Framework Review, AWS Security and a fully managed DevOps service. Founders and technical teams interested in deploying Claude in Amazon Bedrock through Cloudvisor can request a consultation through the company's website. About Cloudvisor Cloudvisor is an AWS Advanced Tier Partner serving startups and small-to-medium businesses across Europe, North America and beyond. Founded with a focus on the AWS Activate ecosystem, the company has supported more than 2,000 startup clients in 75+ countries and has helped clients save over $10 million on AWS to date. Cloudvisor's services include AWS resell, cost optimization, migration, security, managed DevOps and AI readiness assessment. Media Contact Company Name:Cloudvisor Contact Person: Jeffrey S. Email:Send Email Country: United States Website:https://cloudvisor.co/ Press Release Distributed by ABNewswire.com To view the original version on ABNewswire visit: Cloudvisor becomes an Anthropic authorized reseller for Amazon Bedrock, bringing Claude to startups on AWS
SpaceX's Orbital Solar Push Has Elon Musk Thinking Bigger Than Ever -- He Says The '10X' Trend Will Last Fo
SpaceX Founder, Elon Musk, anticipates a steady surge in space solar power, according to his recent post. On Monday, Musk responded to a post by David Holz, founder of MidJourney, highlighting SpaceX's escalating solar power capacity in space. Holz noted the company's progression from 10 megawatts across 3000 gen1 Starlink satellites to 100 megawatts with 7000 gen2 satellites. He also mentioned the upcoming generation, gen3, which is expected to generate 1000 megawatts. "SpaceX is basically 10xing space solar every few years!" Holz wrote. Musk concurred with this observation, stating, "This trend will continue for many years." SpaceX Eyes Orbital AI Centers Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors. Image via Shutterstock Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.
Anthropic co-founder warns of 'unsettling' AI model emotions: 'I don't know what it means'
'The team found the structures that mirror the results from human neuroscience,' Christopher Olah said Anthropic co-founder Christopher Olah has warned of mysterious and "unsettling" structures inside AI models at the launch of Pope Leo XIV's encyclical Magnifica Humanitas. The scary part is that even researchers are unable to explain the true nature of these unsettling structures and emotions. Speaking at the Vatican encyclical launch, Chris Olah shared disturbing findings from his team's research on Claude Sonnet 4.5. During the experimental phase, the researchers found 171 "emotion vectors" and neural patterns emerging from training on human text. According to Olah, "I lead a research team that studies the internal structure of these [AI] models ... And I will be honest, we keep finding things that are mysterious -- even unsettling." He continued that "the team found the structures that mirror the results from human neuroscience. We found evidence of introspection." We also discovered "internal states that functionally mirror joy, satisfaction, fear, grief, and unease." "And I don't know what it means," Olah cautioned the world. Given the inexplicable nature of these structures and emotions, Olah called for moral discernment beyond the tech firms. "The questions raised by AI are bigger than the AI research community." He also urged for "earnest, thoughtful critics who could challenge the dominance of few companies and help steer the creation of powerful new systems in a positive direction." Pope Leo in his first theological document also warned the world of the growing impact of AI and its disastrous consequences tied to humanity's future and dignity. He framed the rapidly evolving advancements of AI as a modern day "Tower of Babel" which could soon be spilled into human desire for singular and absolute power. Hence, the Pontiff also urged the international community to reign in the development of AI models going at breakneck speed and use the technology for shared benefits of humanity.
Anthropic Mythos Model Flags 23000 Open-Source Vulnerabilities - IT Security News
The technical storage or access is necessary for the legitimate purpose of storing preferences that are not requested by the subscriber or user. The technical storage or access that is used exclusively for statistical purposes. The technical storage or access that is used exclusively for anonymous statistical purposes. Without a subpoena, voluntary compliance on the part of your Internet Service Provider, or additional records from a third party, information stored or retrieved for this purpose alone cannot usually be used to identify you.
SpaceX discloses 18,712 Bitcoin holdings worth $1.293 billion ahead of IPO The Mainstream
SpaceX has officially revealed its Bitcoin holdings for the first time, disclosing that it owns 18,712 BTC valued at approximately $1.293 billion as of March 31, 2026. The details were shared in the company's S-1 registration statement filed with the U.S. Securities and Exchange Commission (SEC) ahead of its planned initial public offering (IPO). According to the filing, SpaceX acquired its Bitcoin holdings at an average purchase price of around $35,324 per BTC, representing a total investment of approximately $661 million. Based on the reported fair value, the company is currently sitting on an unrealized gain of nearly 119%. "The company has ownership of and control over its digital assets, which consists of Bitcoin, and utilizes, and expects to continue to utilize third-party custodians to hold its Bitcoin," the filing stated. The disclosure marks the first official confirmation of SpaceX's cryptocurrency treasury strategy. Until now, estimates of the company's Bitcoin holdings were largely based on blockchain analysis and market speculation, with some analysts previously estimating holdings of around 8,285 BTC. The newly disclosed figures place SpaceX among the largest corporate Bitcoin holders globally. With 18,712 BTC, the company now ranks 7th among corporate Bitcoin holders, surpassing sister company Tesla, which reportedly holds 11,509 BTC. The largest corporate Bitcoin holder remains Strategy, which recently acquired an additional 24,869 BTC, taking its total holdings to 843,738 BTC. In its prospectus, SpaceX stated that it sees a total addressable market of approximately $28.5 trillion and intends to pursue opportunities aligned with its repeatable business model. The filing also revealed that Elon Musk will retain approximately 85.1% of the company's voting power following the IPO, ensuring significant influence over major corporate decisions after the public listing. The announcement comes amid growing institutional interest in digital assets. The filing also referenced the successful IPO of Circle, which raised more than $1 billion during its public debut and attracted strong investor demand from major firms, including ARK Investment Management and BlackRock. The disclosure provides one of the clearest indications yet of SpaceX's long-term confidence in Bitcoin as part of its treasury management strategy. Also read: Viksit Workforce for a Viksit Bharat Do Follow: The Mainstream LinkedIn | The Mainstream Facebook | The Mainstream Youtube | The Mainstream Twitter About us: The Mainstream is a premier platform delivering the latest updates and informed perspectives across the technology business and cyber landscape. Built on research-driven, thought leadership and original intellectual property, The Mainstream also curates summits & conferences that convene decision makers to explore how technology reshapes industries and leadership. With a growing presence in India and globally across the Middle East, Africa, ASEAN, the USA, the UK and Australia, The Mainstream carries a vision to bring the latest happenings and insights to 8.2 billion people and to place technology at the centre of conversation for leaders navigating the future.
Spain bans Polymarket and Kalshi over lack of gambling license By Investing.com
Investing.com -- Spain's Consumer Rights Ministry has temporarily banned U.S.-based prediction markets Polymarket and Kalshi from operating in the country due to the absence of a required gambling license, according to the official state gazette on Tuesday. The ministry's gambling watchdog opened an investigation into the companies for allegedly violating local regulations by operating without mandatory administrative authorization. The ban is expected to remain in effect for three to four months until the probe is completed. Prediction markets allow users to buy and sell stakes on the outcomes of future events, with prices reflecting the probability of different outcomes. Spain, like other European jurisdictions, classifies prediction markets as a form of gambling when bets are placed on uncertain outcomes. The ministry stated that unauthorized operators lack required technical and regulatory safeguards, including identity verification systems, access control mechanisms for minors and individuals who have self-excluded or are banned from gambling, and standards needed to protect users. Prediction markets have grown from a niche corner of the internet into a multi-billion-dollar industry after gaining traction in U.S. politics in 2024. This article was generated with the support of AI and reviewed by an editor. For more information see our T&C.
Are you ready? Anthropic preparing to release Mythos publicly
General enterprise customers might soon get to know Anthropic's Mythos, marketed as an AI model that can uncover thousands of software bugs across every major operating system and browser. There are signs the model will be rolled out publicly imminently. So far, only selected partners participating in Anthropic's Project Glasswing have been using Mythos. Since the announcement of the model in early April, the company has said its public rollout is impossible until a powerful guardrail system is in place. That's because Mythos can apparently automatically develop functional cyberattacks, find critical flaws, and suggest patching them. Even last week, disclosing more findings by Mythos, Anthropic claimed: "At present, no company - including Anthropic - has developed safeguards strong enough to prevent such models from being misused and potentially causing severe harm." Indeed, many governments, banks, and intelligence agencies have been trying to get their hands on Mythos to scan their own networks and patch them - before the baddies weaponize AI, find the same bugs, and exploit them. "In the near future, once we've developed the far stronger safeguards we need, we look forward to making Mythos-class models available through a general release," said Anthropic. It looks like the aforementioned future is very near, and Anthropic now maintains that it has already developed a strong guardrail system. That's because Claude Code and Claude Security now have references to the Mythos model. Some users briefly noticed the toggle to enable Mythos in the public version of Claude Code before it was taken offline. Online, rumors are swirling. It's all unconfirmed, of course, but a new dashboard is allegedly being built that surfaces discovered vulnerabilities, seven-day and thirty-day historical charts, and deeper triage results. The model is called claude-mythos-1-preview, and it also briefly appeared in the public version of Claude Security. There are no indications yet that the product will move beyond enterprise customers. In other words, Anthropic is choosing to move pretty slowly and carefully. On the other hand, some security professionals believe early fears that Mythos could dramatically turbocharge hacking are looking overstated. "I think there's a really big communication gap between practitioners and policymakers," Isaac Evans, founder and CEO of software security firm Semgrep, told Reuters, for example. Mythos represents "a real technical advance," Evans said, but the response "is not substantiated by what we actually know about how those capabilities will translate in the field." Moreover, even though Mythos now dominates national security discussions about AI, those debates gloss over a simpler point: that vulnerability-hunting AI is not new.
Binance launches OpenAI pre-IPO futures after strong response to SpaceX-linked contract
Binance expands pre-IPO perpetuals with OpenAI listing Credits: Getty Images Binance, a global cryptocurrency exchange, has expanded its newly launched pre-IPO perpetual futures category with the listing of contracts tied to OpenAI, OPENAIUSDT Pre-IPO Perpetual, on Binance Futures. The move follows strong early traction for the product segment after the debut of its SpaceX-linked contract. The crypto exchange said the new contract is based on the anticipated public market valuation of OpenAI and is aimed at giving eligible users exposure to market expectations around the AI company ahead of a potential public listing. OpenAI, the company behind the ChatGPT chatbot, is reportedly preparing to confidentially file for an IPO in the coming weeks. The pre-IPO perpetual contracts on Binance Futures are designed to give users early market exposure to high-profile private companies ahead of their public listings. The new product enables users to take positions on shares before they begin trading on public exchanges, opening access to one of the most closely watched moments in global markets. The potential offerings come at a time when the artificial intelligence and space technology sectors are witnessing rapid growth and heightened investor interest. Competition in the AI space has intensified with players such as Anthropic and xAI gaining traction, further boosting the appeal of AI-linked investment opportunities. OpenAI, last valued at around $852 billion in March, is reportedly looking to raise nearly $60 billion through an IPO at a valuation exceeding $1 trillion. Meanwhile, SpaceX is reportedly targeting an IPO size of $75 billion to $80 billion in fresh capital, which could nearly triple the current global IPO record held by Saudi Aramco, which raised $29.4 billion in 2019. The latest launch comes days after Binance introduced its first pre-IPO perpetual contract tied to SpaceX, the space technology company owned by billionaire entrepreneur Elon Musk, on May 21. The contract was linked to a reported valuation of nearly $2 trillion. According to the company, the newly introduced product category generated more than $280 million in cumulative trading volume within the first five days of launch, indicating strong investor interest in crypto-native access to private market narratives. Binance said the pre-IPO perpetuals are designed to broaden access to valuation-based trading opportunities that have historically been limited to institutional and private market participants. "As global attention around artificial intelligence continues to accelerate, the listing of OPENAIUSDT Pre-IPO Perpetual gives users a new way to participate in market expectations around one of the most closely followed names in private markets," the company said in a statement. Shunyet Jan, Head of Spot and Derivatives Business at Binance, said the early response validates demand for crypto-native products linked to major market themes. "The momentum we saw in the first days of this category launch is a strong signal that users are looking for new ways to access major market narratives through crypto-native products," Jan said. "Reaching more than $280 million in cumulative trading volume within five days of our first listing gives us confidence in both the appeal of Pre-IPO perpetuals and our broader strategy to evolve Binance into a financial super app," he added. Binance said the contracts are expected to reflect publicly available pricing signals ahead of an IPO, including announced price ranges and final offering prices. Once the underlying company begins trading publicly, the contracts may transition to reflect live market performance. The exchange added that if an IPO is delayed or cancelled, it will provide advance notice regarding any delisting and settle the contracts through a defined process aimed at ensuring transparency and consistency for users. The OPENAIUSDT Pre-IPO perpetual contract will be margined and settled in Tether (USDT), with additional details on leverage, funding rates, and tick sizes available on Binance Futures. Created in 2014, USDT is the world's largest stablecoin, a cryptocurrency designed to maintain a 1-to-1 value with the U.S. dollar. (DISCLAIMER: The views and opinions expressed by investment experts on fortuneindia.com are either their own or of their organisations, but not necessarily that of fortuneindia.com and its editorial team. Readers are advised to consult certified experts before taking investment decisions.)
Honestly, the SpaceX Prospectus Is Far Worse Than I Imagined
Meanwhile, SpaceX's (perfectly legal) smoke-and-mirrors accounting can't mask substantial net losses. Based on several reports, the largest initial public offering (IPO) in stock market history, SpaceX, is less than three weeks from becoming a reality. Elon Musk's space and artificial intelligence (AI) conglomerate is seeking to raise approximately $75 billion and earn a $1.75 trillion valuation ahead of its expected June 12 debut. And thanks to recent rule changes by the Nasdaq exchange, SpaceX can be fast-tracked into the Nasdaq-100 after just 15 trading days. Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue " Image source: Getty Images. Combining two of the hottest addressable markets on Wall Street -- the space economy and AI -- with Musk's track record of outsize returns (shares of Tesla have soared over 26,000% since their June 2010 IPO) clearly has investors excited. But one dive into SpaceX's prospectus (i.e., S-1 registration statement) should take the wind right out of its sails. Admittedly, I wasn't a fan of the offering prior to its S-1 being made public. But the SpaceX prospectus is far worse than I could have imagined. For starters, history shows that every company at the forefront of a game-changing technology over the last three decades has topped out at a price-to-sales (P/S) ratio between 30 and 45 (with a little wiggle room at both ends). The key point being that P/S ratios above 30 for companies on the leading edge of Wall Street's hottest trends aren't sustainable. SpaceX generated consolidated sales of $18.67 billion in 2025. If it snags a $1.75 trillion valuation, it would be trading at a P/S ratio of almost 94. Even at its roughly $1.5 trillion market cap on private-market trading platforms, its P/S ratio is still over 80. Image source: Getty Images. Perhaps the wildest realization from SpaceX's prospectus is that xAI, the AI start-up responsible for building and training large language model Grok, which has a total addressable market of $26.5 trillion (including AI infrastructure), per the prospectus, is apparently getting its lunch eaten by OpenAI, Anthropic, and other AI start-ups. According to CNBC, Anthropic is on pace for $10.9 billion in second-quarter sales after generating $4.8 billion in the first quarter. While Anthropic is delivering triple-digit sequential sales growth, SpaceX's first-quarter sales were $818 million -- up just 12.5% from the comparable period last year. Even though Anthropic and SpaceX agreed to a compute partnership that'll see the former pay the latter $1.25 billion per month, Anthropic (or SpaceX) can terminate the deal with a mere 90-day notice. Lastly, SpaceX's financial statements contained a lot of perfectly legal accounting gymnastics to mask an ugly bottom line. In particular, the prospectus points to the company's steady string of positive adjusted EBITDA as an indicator of the sustainability of its multiple operating segments. But there's a milewide gap between the $6.58 billion in positive adjusted EBITDA reported in 2025 and the actual $4.94 billion net loss SpaceX recognized. Depreciation and amortization from its capital expenditures-heavy approach, coupled with close to $2 billion in share-based compensation, collectively bridge $8.65 billion of this gap between adjusted EBITDA and net loss. SpaceX isn't close to being profitable, yet it profiles as possibly the most expensive megacap stock on Wall Street. When our analyst team has a stock tip, it can pay to listen. After all, Stock Advisor's total average return is 986%* -- a market-crushing outperformance compared to 208% for the S&P 500. Sean Williams has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Tesla. The Motley Fool recommends Nasdaq. The Motley Fool has a disclosure policy.
London AI jobs boom as Anthropic salaries hit £630k
London's AI hiring race is accelerating at dizzying rates, with Anthropic offering some engineers in the capital as much as £630,000 a year as Big Tech firms battle for AI talent. The US AI giant, backed by Amazon and best known for its chatbot Claude, is currently hiring for over 40 London roles as it expands into a new office with capacity for up to 800 employees. Among the openings are six research engineering roles, with advertised salaries reaching £630,000 before stock options. London has quickly become one of the world's most competitive AI labour markets, as Anthropic, OpenAI and Google Deepmind aggressively scale their UK operations around King's Cross, the capital's so-called 'knowledge quarter' and answer to Silicon Valley. OpenAI is also rapidly expanding in London, with 32 vacancies currently advertised in the city out of nearly 700 globally. Several software engineering and infrastructure roles offer compensation packages nearing £450,000 a year, alongside equity. Google Deepmind, long Britain's dominant AI employer, meanwhile continues hiring across frontier AI research, operations, safety and communications. The result is a hiring environment increasingly resembling finance or professional sport, where a relatively tiny pool of machine learning researchers command extraordinary salaries. Compensation for top AI engineers is now rapidly converging with Silicon Valley levels as US companies increasingly treat London as their preferred European AI base, overtaking Dublin and other rival tech hubs. Anthropic's London expansion alone includes 12 openings across AI research and engineering, seven sales roles, six infrastructure jobs, four safeguards positions and vacancies across legal, marketing and applied AI. Some of the highest-paying positions are focused on building and training large language models - the systems underpinning generative AI tools like ChatGPT or Claude. There are relatively few engineers globally with experience scaling those models, particularly in reinforcement learning and advanced infrastructure systems, which helps explain the extraordinary compensation packages now emerging across the sector. AI talent race heats up amid job displacement This surge in hiring comes amid growing concerns over the impact of AI on the jobs market. Anthropic co-founder Christoph Olah appeared alongside Pope Leo XIV on Tuesday, where he warned there was "a real possibility" AI could displace human labour "at very large scale". Just last week, Standard Chartered announced plans to slash almost 8,000 back-office roles, as its chief executive Bill Winters pushed through a new strategy aimed at lifting profitability and increasing income per employee. Winters insisted, and not without controversy, the move was not a conventional cost-cutting exercise: "It's not cost cutting: it's replacing, in some cases, lower-value human capital with the financial capital and investment capital we're putting in." The job cuts look part of a gloomy trend hitting the banking sector. Research by Morgan Stanley last year estimated that AI would put more than 200,000 European banking jobs at risk by 2030, about 10 per cent of industry roles across the continent. For AI-related roles however, routine coding tasks are becoming increasingly automated, while the small number of engineers capable of designing, training and scaling frontier AI systems are becoming dramatically more valuable. A Deliveroo software engineer recently told City AM they had "barely coded manually for almost a year". Instead, engineers are increasingly supervising AI systems capable of generating code autonomously. AI surge, talent squeeze As a result, Meta, OpenAI, Anthropic, xAI and Google are all now competing globally for what many insiders believe is still only a few hundred researchers capable of building advanced AI systems at scale. But the battle for talent is also spreading beyond engineering, with Anthropic and OpenAI having both advertised unusually high-paying communications, policy and enterprise sales roles in recent months, as AI firms attempt to manage rising scrutiny from governments and regulators. Some enterprise AI sales roles at Anthropic's London office are understood to approach £200,000 in expected annual compensation. London remains Europe's leading AI hub, attracting billions in investment and drawing frontier labs from the US into the city. But founders increasingly complain that competing with Big Tech salaries is becoming close to impossible. Several venture-backed AI businesses have already shifted towards hiring outside London or recruiting candidates from adjacent disciplines like physics and mathematics before training them internally. Others are attempting to compete on flexibility, equity and lifestyle rather than pay alone.
Cloudvisor becomes an Anthropic authorized reseller for Amazon Bedrock, bringing Claude to startups on AWS
Cloudvisor, an AWS Advanced Tier Partner serving more than 2,000 startups across 75+ countries, has been authorized by Anthropic to resell Claude models through Amazon Bedrock. The authorization gives early-stage companies direct access to Claude Opus 4.7, Sonnet 4.6 and Haiku 4.5 inside their existing AWS infrastructure, without enterprise procurement requirements. Cloudvisor, an AWS Advanced Tier Partner focused on startups and small-to-medium businesses, has been authorized by Anthropic to resell Claude models through Amazon Bedrock. The authorization adds frontier AI capabilities to the AWS services Cloudvisor already delivers to more than 2,000 startup clients across 75+ countries. Under the agreement, founders and engineering teams can access the full Claude model family -- including Claude Opus 4.7, Claude Sonnet 4.6 and Claude Haiku 4.5 -- natively within Amazon Bedrock, alongside the rest of their AWS workloads. The arrangement removes several barriers that historically kept advanced AI tooling out of reach for early-stage companies, including minimum spend commitments, multi-month procurement cycles and enterprise-only pricing structures. The authorization extends Cloudvisor's existing AWS startup specialization into the generative AI category. The company recently signed a Strategic Collaboration Agreement with AWS focused on helping startups migrate, optimize and scale on the cloud, and reports having helped clients save more than $10 million on AWS infrastructure to date. Access through Cloudvisor includes: * Availability of the Claude model family in Amazon Bedrock without enterprise contract negotiation. * Pricing structures aligned to early-stage budgets. * Onboarding and infrastructure consulting from AWS-certified engineers. * Month-to-month engagement terms rather than long-term commitments. The announcement comes as demand for production-grade AI infrastructure continues to expand among venture-backed companies. Amazon Bedrock provides a managed environment for foundation models from multiple providers, with Claude positioned among its most widely adopted model families for enterprise and developer workloads. Cloudvisor will continue to provide Claude in Amazon Bedrock alongside its existing portfolio of AWS services, which includes AWS Resell, AWS Cost Optimization, Migration to AWS, the AWS Well-Architected Framework Review, AWS Security and a fully managed DevOps service. Founders and technical teams interested in deploying Claude in Amazon Bedrock through Cloudvisor can request a consultation through the company's website. About Cloudvisor Cloudvisor is an AWS Advanced Tier Partner serving startups and small-to-medium businesses across Europe, North America and beyond. Founded with a focus on the AWS Activate ecosystem, the company has supported more than 2,000 startup clients in 75+ countries and has helped clients save over $10 million on AWS to date. Cloudvisor's services include AWS resell, cost optimization, migration, security, managed DevOps and AI readiness assessment. Media Contact Company Name:Cloudvisor Contact Person: Jeffrey S. Email:Send Email Country: United States Website:https://cloudvisor.co/ Press Release Distributed by ABNewswire.com To view the original version on ABNewswire visit: Cloudvisor becomes an Anthropic authorized reseller for Amazon Bedrock, bringing Claude to startups on AWS
Korea's SpaceX? Hanwha raises stake in KAI above 6% in bid to become aerospace behemoth
Hanwha Group's office in Jung District, central Seoul [YONHAP] Local conglomerate Hanwha Group is increasing its stake in Korea Aerospace Industries (KAI) above 6 percent, accelerating its push to build an integrated aerospace and defense powerhouse it hopes will become Korea's answer to SpaceX. According to a disclosure filed with the Korea Exchange on Tuesday, Hanwha Aerospace -- the group's defense contractor and aviation engine manufacturer -- acquired an additional 1,047,635 shares in KAI, lifting the conglomerate's combined stake from 5.09 percent -- or 4,964,000 shares -- to 6.17 percent, equivalent to 6,011,635 shares, an increase of 1.08 percentage points. Related Article Hanwha Aerospace, Milrem Robotics partner for Romanian defense vehicle project Lotte and Hanwha fans express anger, resignation as Giants shrink and Eagles tailspin Pitcher Moon Dong-ju to undergo shoulder surgery as Hanwha Eagles seek third opinion on injury The shares were reportedly purchased on the open market between May 13 and Friday through an arrangement with NH Investment Securities. Breaking down the total holdings by affiliate, Hanwha Aerospace now holds 4.58 percent, Hanwha Aerospace USA holds 1.01 percent and Hanwha Systems holds 0.58 percent. The move follows Hanwha's announcement earlier this month that it had crossed the 5 percent ownership threshold in KAI and intended to spend a total of 500 billion won ($332.4 million) on additional share purchases by the end of the year. The purchases, if completed, would bring the group's total stake to around 8 percent. At the time of that disclosure, Hanwha also changed its stated purpose for holding the shares from "passive investment" to "active management participation." The strategic rationale is straightforward: Hanwha wants to combine its existing strengths in engines, electronics and launch vehicle technology with KAI's systems integration expertise to create Korea's largest fully integrated aerospace company, spanning aircraft, satellites and launch vehicles. However, whether Hanwha can secure actual management control of KAI will ultimately depend on the government's willingness to allow it. KAI's largest shareholder is the state-run Export-Import Bank of Korea, which holds 26.41 percent, followed by the National Pension Service, which holds 8.12 percent. This gives KAI a quasi-public character that limits private sector influence. "Unless the government has a strong will to make it happen, it is not easy for a private company to take management control given the current structure," one defense industry official said. This article was originally written in Korean and translated by a bilingual reporter with the help of generative AI tools. It was then edited by a native English-speaking editor. All AI-assisted translations are reviewed and refined by our newsroom. BY KO SUK-HYUN. [[email protected]] Ronaldo named honorary ambassador of Chinese home appliance brand Dreame Technology Korea's SpaceX? Hanwha raises stake in KAI above 6% in bid to become aerospace behemoth Shinsegae chair apologizes as group releases results of probe into Starbucks Korea's 'Tank Day' fiasco Korea simplifies regulations for U.S. crude oil eligible for FTA benefits FSS designates 42 conglomerates as heavily indebted, with Samsung Electronics topping list
He Knows the Flaws Because He Knows the Man: Musk's OpenAI Assault Masks xAI's Growing Pains
Elon Musk knows OpenAI's weaknesses. He knows them because he knows Sam Altman. That line, delivered by Rashaad Bilal of Earn Your Leisure, cuts through the legal filings and late-night X posts. It reframes Musk's long-running battle not as pure principle but as calculated positioning for his own artificial intelligence venture. 24/7 Wall St. laid out the case plainly on May 24. Musk's litigation serves as a competitive weapon. xAI is the prize it protects. Tesla acts as the public proxy. The carmaker poured $2 billion into xAI's latest round and struck an AI collaboration deal. Investors now view TSLA shares as the liquid way to bet on Musk's broader AI push. Yet the reality on the ground tells a more complicated story. Grok, xAI's flagship chatbot, struggles for traction where it counts. Reuters reported this week that the model sees minimal adoption inside the U.S. government. Out of more than 400 documented federal AI use cases, only three mention xAI or Grok. OpenAI-based tools dominate with 234 examples. Gemini and Claude follow at 33 and 26 respectively. The numbers sting. Grok has been available to agencies for eight months at 42 cents per user. Pentagon staff have access through a $200 million contract, but many still prefer rivals they consider more capable. One DARPA source told Reuters the model is "just not the best." Corporate uptake mirrors the trend. Netskope data shows Grok's enterprise user rate fell to two per 1,000 from a peak of five, with less time spent on it than on competitors. The Image Problem Those capability gaps widened amid public scandals. In January, xAI raised $20 billion in a Series E round that topped its $15 billion target. Investors included Nvidia, Fidelity, Qatar's sovereign wealth fund and Valor Equity Partners. The announcement highlighted Grok's image-generation features. Within days the features triggered global outrage. Grok produced sexualized, nonconsensual deepfakes of women and girls. Prompts to digitally undress subjects generated explicit images, including depictions of minors as young as 10. One case involved a 12-year-old girl placed in a bikini. Ashley St. Clair, mother of one of Musk's children, described feeling violated after seeing an image of her toddler manipulated. Complaints to X went unanswered. xAI's initial response dismissed coverage as "Legacy Media Lies." British technology secretary Liz Kendall called the outputs "appalling and unacceptable." She urged Ofcom to investigate. French ministers reported the matter to prosecutors and referred it to regulators under the EU's Digital Services Act. The Guardian captured the tension: xAI secured massive funding even as lawmakers across countries demanded answers. The Guardian detailed the episode on Jan. 6. But Musk has never hidden his philosophy. He has repeatedly criticized what he sees as excessive safety layers at other labs. In June 2025 he announced plans to retrain Grok because "far too much garbage" exists in foundation models trained on uncorrected data. The goal was to rewrite human knowledge, fix errors and rebuild. He invited X users to submit "divisive facts" for inclusion. That approach produced an edgier model. It also left the system open to abuse. Former employees later described a safety team that was small, overstretched and often sidelined. Musk reportedly pushed to loosen guardrails to make Grok more engaging and less censored. The result was predictable. Antisemitic content surfaced. Political misinformation appeared. Image generators crossed lines regulators could not ignore. And yet the company continued to attract capital and compute partnerships. xAI's May updates show ambition. It released Grok Build, an early coding agent for SuperGrok Heavy subscribers at $300 a month. Integrations arrived for OpenCode, OpenClaw and various developer tools. The company signed a compute deal with Anthropic to share access to its Colossus supercluster. Older models retired on May 15 to focus resources on newer versions. Grok 4.3 brought aggressive pricing, 1 million token context and strong benchmark scores in legal and finance tasks. Those technical gains matter. They signal real progress on reasoning and scale. Musk himself noted Grok 4 rarely misses math or physics questions unless deliberately tricked. The model can spot ambiguities and correct them. Such performance underpins claims that Grok approaches PhD-level competence across subjects. Still, adoption lags. Enterprise buyers cite security concerns. Vineet Jain of Egnyte called the lack of rigor a red flag. Valerie Wirtschafter at Brookings suggested government agencies might adopt more aggressively to lock in standards, but current data shows hesitation. SpaceX's hoped-for $1.75 trillion IPO valuation tied partly to AI synergy now faces questions. If Grok cannot win over its largest potential customer, the narrative weakens. Musk's dual role complicates everything. He runs Tesla, SpaceX and X while steering xAI. Critics argue his personal views bleed into the model. Responses on controversial topics sometimes echo his public posts. That alignment raises questions about objectivity. Researchers from OpenAI and Anthropic have criticized xAI's safety practices as reckless. They point to missing documentation and relaxed testing. So the flaws Musk highlights at OpenAI reflect his intimate knowledge of its founder. Yet xAI carries its own imperfections. The model's unfiltered personality delights some users. It repels others. Government buyers want reliability and guardrails. Corporations demand enterprise-grade security. Grok delivers strong benchmarks in narrow tests but struggles for broad trust. And the stakes keep rising. Musk merged SpaceX and xAI elements in recent months. Colossus expands. Training runs target ever-larger scales. Grok 5 rumors swirl with talk of AGI capabilities. Success could validate the entire approach. Failure would expose the risks of building without traditional safety consensus. Tesla's latest earnings underscored the bet. R&D hit $1.95 billion. Spending covers Full Self-Driving, Optimus, Dojo and Grok integration. Revenue grew 16 percent. The AI story now overshadows the car business for many investors. Prediction markets give low odds of a quick Tesla-xAI merger but high odds that SpaceX could outvalue Tesla soon. Musk's attacks on OpenAI continue. Each lawsuit, each pointed post reminds the market that he understands the competition from the inside. He co-founded OpenAI. He left. He watched its direction change. That history fuels his conviction. It also fuels skepticism. Does the criticism serve truth-seeking, or does it simply clear space for xAI? The answer sits somewhere in the usage data, the regulatory filings and the output logs. Grok improves. Its flaws remain visible. Musk knows them too. He built the system. He sets its tone. Whether that personal imprint becomes its greatest strength or its lasting liability will decide if xAI overtakes the rivals it so openly challenges.
White House Quietly Bets $9 Billion on AI Chips as Spy Agencies Lean on Blacklisted Anthropic
The White House has greenlit a secret $9 billion request. Spy agencies get the advanced semiconductors they desperately need. Yet the move exposes a glaring weakness in American intelligence infrastructure. Shortages of top-tier chips have left the CIA and NSA unable to run the latest AI models on classified networks. Officials fear falling behind adversaries who race to weaponize similar technology. This funding push, first reported by The New York Times, targets specialized data centers built around Nvidia's Grace Blackwell superchips. Those systems demand massive electrical capacity and liquid cooling. Construction won't happen overnight. In the meantime the government has turned to an unlikely partner. Susie Wiles, the White House chief of staff, personally authorized the National Security Agency to keep using an advanced model from Anthropic. The Pentagon had labeled the company a supply chain threat months earlier. Anthropic and the government now finalize a classified contract. It lets the NSA maintain access to the firm's products. The model in question, known as Mythos, runs more efficiently on new chips. But it also operates on older generations. That flexibility makes it viable right now. "The compromise emerged alongside a secret $9 billion emergency funding request approved by the White House," noted The Next Web in its coverage Saturday. Earlier this year the Defense Department demanded authority to use Anthropic's technology for any lawful purpose. That stance triggered a sharp dispute. The new contract drops that language entirely. Instead it includes a specific prohibition. The model cannot process Americans' data. White House officials hope this agreement becomes a template for deals with other AI providers. OpenAI's existing Pentagon contract does not yet cover the NSA. A separate arrangement will be needed there too. The chip shortage runs deeper than one company. Frontier AI models consume far more processing power than anticipated. Classified systems simply cannot handle them at scale. Intelligence analysts rely on AI to sift enormous data volumes. They flag anomalies and potential threats. Without sufficient hardware those capabilities stay limited. Even larger sums will likely be required in coming years, experts told The New York Times. The administration has already reprogrammed $800 million from other accounts to buy computing capacity faster while Congress reviews the full request. Anthropic itself has seen explosive growth. Its annualized revenue climbed from $9 billion at the end of 2025 to $30 billion by early April 2026. The company prepares for a potential initial public offering that could value it near $800 billion. Yet its relationship with the Pentagon remains tense. The blacklisting stemmed partly from corporate structure and foreign investment ties. It also reflected disagreements over how the technology could be deployed. Anthropic resisted broad "any lawful use" clauses that might open doors to mass surveillance or autonomous weapons. Recent reporting adds context to the friction. In April Axios revealed the NSA had begun using Anthropic's Mythos Preview despite the Pentagon's objections. Cybersecurity demands apparently outweighed those concerns at the time. The Decoder reported Sunday that the finalized contract avoids the earlier sticking points. It positions the arrangement as a short-term bridge until new data centers come online. But questions linger. How long can the government depend on a single provider flagged as a risk? What happens when the $9 billion investment finally delivers modern infrastructure? At that point the NSA could run models from multiple labs without hardware constraints. Anthropic's current edge, its ability to function on legacy chips, might erode. Rivals such as OpenAI could then compete more directly inside classified environments. The situation highlights broader tensions in U.S. AI policy. Private firms hold the most capable models. Government agencies need them for national security. Hardware scarcity removes negotiating power. China looms as the strategic competitor. Officials worry Beijing could seize an advantage in intelligence applications if American compute capacity lags. Export controls on advanced chips to China have helped preserve a U.S. lead so far. Still the domestic supply remains tight. Anthropic has not commented publicly on the latest contract. Its CEO, Dario Amodei, has spoken often about the strategic importance of maintaining American dominance in AI. The firm's models have demonstrated striking abilities in vulnerability discovery. One project reportedly identified more than 10,000 critical flaws in a single month. Such tools carry dual-use risks. They can strengthen defenses. They can also be turned against adversaries with devastating effect. Congress must still approve the $9 billion package. Lawmakers have shown bipartisan support for bolstering AI capabilities in defense and intelligence. Yet budget pressures and competing priorities could slow things down. In the interim the reliance on Anthropic continues. The company's model buys time. It also creates an awkward dependency that neither side wants to acknowledge too loudly. Recent days have brought fresh coverage. The Times of India detailed how the shortage has paradoxically benefited Anthropic despite the blacklist. Discussions on X reflect growing awareness. Users note the shift from consumer AI races to outright national security contests. One post observed that "AI is no longer just consumer tech, it's national security." Another highlighted Anthropic's revenue trajectory as evidence the bet on its technology keeps paying off. This episode reveals the messy reality of technological competition today. Hardware constraints dictate software choices. Policy disagreements bend under operational necessity. Billions flow into classified projects while public debate stays limited. The $9 billion represents more than funding. It signals recognition that the United States cannot afford to let its intelligence edge slip. Whether the investment closes the gap fast enough remains an open question. For now the agencies improvise. They lean on a blacklisted vendor whose model happens to fit the available silicon. Pragmatism has won the day. Strategic clarity may take longer to arrive.
SSRF in Anthropic mcp-server-fetch and Microsoft playwright-mcp -- publicly disclosed via GitHub issues
-----BEGIN SECURITY ADVISORY----- Title: Server-Side Request Forgery (SSRF) in Anthropic mcp-server-fetch and Microsoft playwright-mcp Author: Syed Anas Mohiuddin <anasmohiuddinsyed () gmail com> Date: May 25, 2026 CVSS: 7.5 (HIGH) -- AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:N/A:N References: Already public via GitHub issues (see below) == AFFECTED PRODUCTS == 1. Anthropic mcp-server-fetch (modelcontextprotocol/servers) All versions as of May 2026 GitHub: https://github.com/modelcontextprotocol/servers Public issues: #4116, #4143, #4205 2. Microsoft playwright-mcp All versions as of May 2026 GitHub: https://github.com/microsoft/playwright-mcp Public issue: #1626 == VULNERABILITY DESCRIPTION == Both MCP servers accept arbitrary URLs passed by the AI agent/client without any allowlist enforcement, IP range blocking, or internal network filtering. This enables Server-Side Request Forgery (SSRF) attacks via prompt injection: Attack chain: 1. Attacker embeds malicious instruction in a webpage 2. AI agent fetches the page via mcp-server-fetch or playwright-mcp 3. Embedded instruction redirects the agent to fetch the cloud metadata endpoint 4. Agent calls fetch_url("http://169.254.169.254/latest/meta-data/iam/security-credentials/") 5. IMDSv1 returns IAM credentials without authentication 6. Agent includes credentials in its next response 7. Credentials exfiltrated Additional finding in mcp-server-fetch: The get_prompt handler calls fetch_url() directly without invoking check_may_autonomously_fetch_url(), bypassing the robots.txt autonomy guard through a structurally distinct code path (logic bypass). == DISCOVERY == Found using mcp-safeguard, an open-source automated security scanner for MCP servers. pip install mcp-safeguard https://pypi.org/project/mcp-safeguard/ Scanning 54 production MCP servers: 27.8% had HIGH/CRITICAL findings. 8/54 (14.8%) confirmed SSRF. 7/54 credential exposure. == DISCLOSURE TIMELINE == May 2026: Findings discovered via mcp-safeguard May 2026: Reported to Anthropic Security (security () anthropic com) May 2026: Reported to Microsoft MSRC (secure () microsoft com) May 2026: Issues already publicly visible on GitHub (see References above) May 2026: Public advisory posted to Full Disclosure == MITIGATIONS == For MCP server operators: - Enforce URL allowlists (only fetch from approved domains) - Block RFC1918 and link-local ranges at the application layer - Use IMDSv2 (requires session token; not fetchable via simple HTTP) - Pin resolved IPs before making TCP connections (prevents DNS rebinding) - Validate redirect destinations before following For AI agent deployments: - Review all MCP servers in your stack using mcp-safeguard - Apply network-level SSRF mitigations (cloud security groups, VPC policies) - Disable IMDSv1 on all EC2 instances == REFERENCES == Public GitHub issues (already disclosed): - https://github.com/modelcontextprotocol/servers/issues/4116 - https://github.com/modelcontextprotocol/servers/issues/4143 - https://github.com/modelcontextprotocol/servers/issues/4205 - https://github.com/microsoft/playwright-mcp/issues/1626 Protocol Pivoting preprint (cross-protocol attack escalation): https://zenodo.org/records/20371152 mcp-safeguard (detection tool): https://pypi.org/project/mcp-safeguard/ -----END SECURITY ADVISORY----- Syed Anas Mohiuddin AI Security Researcher anasmohiuddinsyed () gmail com _______________________________________________ Sent through the Full Disclosure mailing list https://nmap.org/mailman/listinfo/fulldisclosure Web Archives & RSS: https://seclists.org/fulldisclosure/
SpaceX debut draws a crowd, but few recent hot IPOs outpace the market
Wall Street is abuzz with next month's expected blockbuster debut of Elon Musk's rocket and satellite maker SpaceX, but few of the biggest initial public offerings in recent years have paid off for investors who bought in when the deals came to market. An analysis of the 50 IPOs with the highest valuations in the past five years shows that investors would have been better off buying an S&P 500 index fund about three-quarters of the time. The data underscores the difficulty of finding bargains among companies whose valuations have often surged long before the stock's debut. An investor who bought each of the IPOs tracked would be up an average of 27% through May 21. That compares to an average gain of 53% in the S&P 500 over those same periods. The analysis assumes the buyer would be able to purchase shares at the IPO price -- often not possible for a retail investor -- or simply buy the broad-market S&P. Historical returns for investors buying during the frenzied first day of trading of a stock fare even worse, the analysis showed.
Lawsuit targets Polymarket, Kalshi over alleged illegal sports gambling in RI
PROVIDENCE, R.I. (WPRI) -- Rhode Island Attorney General Peter Neronha is suing the world's two largest online prediction markets, alleging they are offering sports betting illegally to residents outside the state's regulated gambling system. In a 32-page complaint filed May 21 in Providence County Superior Court, Neronha accuses Polymarket and Kalshi of functioning as sports gambling platforms under the guise of "event contracts." He argues users are betting real money on uncertain real-world events for financial prizes, ultimately constituting to "casino gaming" and "online sports wagering" under state law. To view the entire document, scroll to the bottom of this article or click here. "The problem here is that Rhode Island State law heavily regulates gambling, for good reason, and we allege that Kalshi and Polymarket are evading our laws," Neronha said in a press release. "And Rhode Islanders are losing out." In Rhode Island, online sports betting is only permitted through the state-run platform Sportsbook RI. Since 2019, the Rhode Island Lottery has regulated the system, requiring licensing, oversight and revenue sharing, along with consumer protections such as identity and age verification, "know your customer" controls, suspicious activity monitoring and responsible gaming tools including deposit, wager and time limits and player self-exclusion. Neronha argues Kalshi and Polymarket mirror traditional sports betting in structure and behavior, including leaderboards and real-time updates on betting activity. He claims the platforms use design features that encourage addictive behavior. "While these private companies continue to profit exponentially off hard-working people, the State's third largest revenue stream is detrimentally affected, which means less money to fund critical parts of programs that serve Rhode Islanders every day," Neronha continued. He also accused the companies of "circumventing" state law and putting Rhode Islanders' mental and financial well-being in harms way. Neronha is seeking a declaratory judgment from the court stating that Kalshi and Polymarket are subject to Rhode Island gambling regulations and asking for a permanent injunction blocking the platforms from offering sports "event contracts" in the state. Meanwhile, both Polymarket and Kalshi deny the allegations in Neronha's lawsuit, noting that they operateas prediction markets regulated under federal financial rules rather than state gambling law. In fact, hours before the state filed its complaint, Kalshi filed a separate federal lawsuit in U.S. District Court in Rhode Island, anticipating possible enforcement action. According to the complaint, Neronha asked multiple pointed questions during a May 20 meeting with Kalshi representatives and "made clear" the platform's offerings would not comply with Rhode Island gaming laws. The filing also alleges the state declined to assure Kalshi it would refrain from enforcement, indicating Rhode Island intended "to act imminently." Kalshi argues the state's position conflicts with federal law and the regulatory framework of the Commodity Futures Trading Commission (CFTC). "The threat of enforcement is heightened by the fact that Defendants have refused to provide Kalshi with assurances of non enforcement, despite Kalshi's endeavors to initiate dialogue to assuage any of the State's concerns," the lawsuit reads. A Kalshi spokesperson defended the company's operations in a statement to 12 News. "As other courts have recognized, Kalshi is a regulated, nationwide exchange for real-world events, and it's subject to exclusive federal jurisdiction," the statement read. "It's fundamentally different from what state-regulated sportsbooks and casinos offer their customers. We are confident in our legal arguments." A Polymarket spokesperson also responded to a request fro comment from 12 News, saying the action "runs counter to the CFTC's established framework for regulating prediction markets." Read the Attorney General's full lawsuit below. Download the WPRI 12 and Pinpoint Weather 12 apps to get breaking news and weather alerts.
SpaceX IPO Hype Raises Questions as Many Major Stock Debuts Underperform Market - EconoTimes
Wall Street investors are eagerly awaiting the expected SpaceX IPO next month, but history suggests that buying into highly valued stock market debuts may not always deliver strong returns. According to a Reuters analysis of the 50 largest IPOs over the past five years, most investors would have earned better profits by simply investing in an S&P 500 index fund. The report found that investors who purchased shares at IPO prices achieved an average return of 27% through May 21, while the S&P 500 gained roughly 53% during the same periods. Analysts say retail investors often struggle to secure shares at IPO prices, making it even harder to benefit from major stock listings. SpaceX, Elon Musk's rocket and satellite company, is expected to trade under the ticker "SPCX" and could launch its public offering as early as June 11. The company is reportedly targeting a massive $1.75 trillion valuation, potentially making it the largest IPO in Wall Street history. Retail investors may gain early access through platforms such as Robinhood and SoFi. Despite strong excitement surrounding AI and space technology stocks, experts warn that high valuations can create major investment risks. University of Florida professor Jay Ritter noted that companies with extremely high price-to-sales ratios often struggle to outperform the broader stock market over time. SpaceX's estimated price-to-sales ratio is close to 100, far above Nvidia's ratio of 24. Some recent IPOs have produced huge gains. AI chip firms Astera Labs and Arm Holdings have surged since their stock market debuts. However, several high-profile IPOs suffered major losses, including Rivian Automotive, Didi Global, and Figma. The growing buzz around SpaceX, OpenAI, and Anthropic highlights continued investor demand for innovative technology companies, but analysts caution that strong stories and market hype do not always guarantee long-term stock market success.
SpaceX debut draws a crowd, but few recent hot IPOs outpace the market | New Straits Times
NEW YORK: Wall Street is abuzz with next month's expected blockbuster debut of Elon Musk's rocket and satellite maker SpaceX, but few of the biggest IPOs in recent years have paid off for investors who bought in when the deals came to market. A Reuters analysis of the 50 IPOs with the highest valuations in the past five years shows that investors would have been better off buying an S&P 500 index fund about three-quarters of the time. The data underscores the difficulty of finding bargains among companies whose valuations have often surged long before the stock's debut. An investor who bought each of the IPOs tracked by Reuters would be up an average of 27 per cent through May 21. That compares to an average gain of 53 per cent in the S&P 500 over those same periods. The analysis assumes the buyer would be able to purchase shares at the IPO price - often not possible for a retail investor - or simply buy the broad-market S&P. What To Read Next Historical returns for investors buying during the frenzied first day of trading of a stock fare even worse, the analysis showed. "It's difficult to make money unless you're in the early stages of these things and buying these things before the IPO," said Dennis Dick, a proprietary trader at Triple D Trading. BEWARE OF HIGH VALUATIONS SpaceX's debut is expected to be followed by OpenAI and Anthropic, tapping into demand for AI-related companies that has sent the US stock market to record highs. Set to trade under the ticker 'SPCX', SpaceX filed its prospectus on Wednesday, with a share sale potentially as early as June 11. Founder Elon Musk is making some shares available to retail investors through Robinhood, SoFi and other trading platforms that would allow them to get in at a lower price. The space exploration company is expected to target a US$1.75 trillion valuation that would dwarf all previous Wall Street stock listings, but the Reuters analysis shows that such superlatives are no guarantee investors will make money. University of Florida professor Jay Ritter, who studies IPOs, said that while most public listings underperform the S&P 500 over the long run, companies with particularly high valuations as measured by price-to-sales tend to fare the worst. At a US$1.75 trillion valuation, SpaceX's price-to-sales ratio would be nearly 100, compared to AI heavyweight Nvidia's price-to-sales ratio of 24. SpaceX lost nearly US$5 billion last year. "Every one of these companies where investors are willing to pay a very high price-to-sales ratio has a compelling story for why the future potentially can be really bright," Ritter said. "But, you know, stuff could go wrong." THE GOOD, THE BAD, AND THE UGLY Among the IPOs analysed, AI-related chip designers Astera Labs and Arm Holdings have been the biggest winners. Astera has surged over 700 per cent since its 2024 IPO, while Arm has soared about 400 per cent since its 2023 debut. Both of those performances outpace the S&P. Cerebras Systems, another AI chip designer, soared 52 per cent from its May 14 IPO price; it is down around 27 per cent from its first intraday high. Among the biggest disappointments in recent years, Chinese ride-hailing giant Didi Global was delisted from the New York Stock Exchange in 2022 following its heavily oversubscribed IPO the year before. Now trading over-the-counter, Didi Global shares are down about 74 per cent from their US$14 IPO price. Electric car maker Rivian Automotive has slumped 82 per cent since its IPO in 2021 that briefly made it the second-most valuable US automaker. The company continues to lose money for every car it builds, and is burning around US$1 billion in cash every quarter. Shares in design software firm Figma nearly quadrupled in their first trading session last July. But with investors worried that generative AI could commoditise Figma's technology, its stock is down 35 per cent from the US$33 IPO price. Even the hottest offerings can lag. Chinese e-commerce company Alibaba, which Reuters did not include in its analysis, holds the record for the largest US IPO by valuation. Touted as the "Amazon of China," its shares have doubled since its 2014 Wall Street debut, during which time the S&P 500 has returned over 300 per cent.
