News & Updates
The latest news and updates from companies in the WLTH portfolio.
CoreWeave stock soars 13% on Anthropic deal
CoreWeave (CRWV) stock rocketed 13% higher on Friday after the company announced it is entering a multiyear agreement with Anthropic (ANTH.PVT) that will see the AI cloud company provide Anthropic with computing capabilities to build and power its AI models. CoreWeave said Anthropic will use its cloud services to run workloads at "production scale," and that it will initially focus on a phased rollout with the option to expand the agreement in the future. The companies didn't provide the terms of the deal, including pricing or how many gigawatts of chips it will cover. The announcement comes after Reuters reported that Anthropic is also considering designing its own semiconductors to contend with the AI chip crunch. Earlier this week, Anthropic also said it's working with Broadcom (AVGO) and Google to use 3.5 gigawatts of Google's Broadcom-made Tensor Processing Units. AI companies across the board are working to secure as many semiconductors as possible as they build out their AI services. Anthropic rival OpenAI (OPAI.PVT) is also developing its own chips. In October, the company entered into a partnership with Broadcom to develop upwards of 10 gigawatts of custom semiconductors for its various AI services. That's in addition to deals it has with both Nvidia (NVDA) and AMD (AMD). And last month, Meta (META) revealed four new custom AI processors, including its MTIA 400, which the company said delivers raw performance rivaling some of the top chips on the market. The social media giant, like Anthropic, also entered into a deal with CoreWeave that will see CoreWeave power Meta's AI services through December 2032. CoreWeave said the capacity will be spread out among a number of its data center locations and include some of the first deployments of Nvidia's upcoming Vera Rubin system. In January, Microsoft (MSFT) also revealed a new custom AI chip that will serve as an alternative to Nvidia and AMD's offerings. Amazon (AMZN) and Google (GOOG, GOOGL) have been using their own chips for years. Unlike Microsoft, however, those companies are looking to sell or rent their chips to third-party customers. In February, The Information reported that Meta inked a deal with Google to rent that company's TPUs and is exploring purchasing them for its own data centers. Amazon CEO Andy Jassy also pitched the idea of selling the company's chips in large servers to third-party customers in his latest annual shareholder newsletter on Thursday. Email Daniel Howley at [email protected]. Follow him on Twitter at @DanielHowley. For the latest earnings reports and analysis, earnings whispers and expectations, and company earnings news, click here
Musk's SpaceX Reportedly Suffered $5 Billion Loss Last Year Despite High Revenue
Elon Musk's SpaceX reportedly recorded a loss of nearly $5 billion last year, despite raking in over $18.5 billion in revenue. The loss incurred by SpaceX includes figures from xAI, an artificial intelligence firm founded by Musk and acquired by SpaceX in February 2026. The financial data, which consolidates the performance of both SpaceX and xAI, has only recently been disclosed, according to The Information on Thursday. SpaceX did not immediately respond to Benzinga's request for comment. SpaceX AI Bet, IPO Concerns Grow 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.
UniPat AI Launches EchoZ Prediction Model, Demonstrating Performance Beyond Human Traders on Polymarket
Prediction markets are a place where individuals can trade on the results of a future event. Third-party data, including from Dune Analytics, show diverse participant performance over time in such liquid (and competitive) markets. The model was tested against scenarios with multiple variables, the company said, which included regulatory developments, geopolitical occurrences, on-chain governance decisions and market-related events. These scenarios were chosen to represent situations with partial or evolving information. UniPat AI has also performed a set of stress tests regarding its evaluation framework. These modifications consisted of changing scoring parameters, partially removing input data, and modifying model configurations. Under these conditions, models maintained consistent rankings within the company's testing framework, it said. UniPat AI claims it has released prediction inputs, probability estimates, timestamps and ultimate outcomes to the public for independent auditing. It consists of aligned and non-aligned predictions. Examples it cites include forecasts for equity market capitalisation, digital asset price levels and professional sports standings. Probability estimates were documented before outcome resolution in each case for validation. EchoZ-1. 0 relies on a method named "Train-on-Future," UniPat AI said, and differs from methods that mainly use historical data. It includes evaluation criteria to capture reasoning structure as well as outcome alignment. In this process, a system called Automated Rubric Search is used to extract reasoning patterns related to the observed actions. This development team consists of researchers with expertise in reinforcement learning, data synthesis, and model evaluation. The prediction markets were chosen as a testing ground because of their measurable and verifiable nature, UniPat AI said. A spokesman for the company said structured assessments of probabilities may have applications in places where there's decision-making under uncertainty. About UniPat AI UniPat AI is a research-focused artificial intelligence company working on the development of machine learning systems designed for real-world applications. The organisation describes its mission as advancing AI systems from experimental models to practical, deployable tools capable of operating in complex environments. Media Contact Organization: UniPat AI Contact Person Name: Yao He Website: https://unipat.ai/ Email: [email protected] Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.
Irish army on standby as widespread fuel protests cause chaos across country
Add Yahoo as a preferred source to see more of our stories on Google. Ireland is teetering on the "precipice of turning oil away from the country," Premier Micheal Martin has warned, as widespread protests over soaring fuel costs continue to disrupt supply chains. The Taoiseach condemned the "unconscionable" blockading of ports and the nation's sole oil refinery by demonstrators. Yet, protest leader John Dallon vowed the action would persist "until we get a result," stating he was refused entry to a government meeting with representative bodies on Friday. This escalating disruption has prompted gardai to attend a demonstration at the Cork refinery, amid concerns that Irish forecourts risk running dry. Days-long blockades of major supply depots, sparked by rising fuel prices linked to the war in Iran, have caused the shortages. While protests continued on Friday, activists reported that "more fuel" deliveries were now being permitted. Hauliers and agricultural contractors began a series of protests on Tuesday, which have grown from slow-moving convoys on motorways and restricted access to Dublin's busiest streets, to blockades of fuel depots that supply half the country. Access has been restricted to Ireland's oil refinery in Whitegate, Co Cork, as well as fuel depots in areas such as Galway City and Foynes, Co Limerick. It has raised concerns over panic buying at forecourts, some of which have run out of fuel, as well as impacts on emergency services and deliveries of key supplies for animal welfare on farms. Government leaders have condemned the protests as "wrong", and said the fuel blockades were holding the country to "ransom". The Irish army remains on standby to help gardai remove blockades at critical infrastructure, which includes fuel depots. Gardai have confirmed they are "at the scene of an incident in Whitegate, Co Cork" on Friday morning. A spokesperson for protesters at Foynes Port said they would allow feed and five loads of fuel to be transported to frontline workers and hospitals, after allowing chemicals needed by Uisce Eireann through their blockade on Thursday. Mr Martin told RTE: "We are now on the precipice of turning oil away from the country in the middle of a global oil supply." He said "somebody else will buy" oil that is on a tanker off the coast of Galway if it cannot dock. Asked if Ireland is at risk of "losing all its oil supply", Mr Martin replied, "correct". He added: "It is unconscionable, it's illogical, it is difficult to comprehend." Transport Minister Darragh O'Brien said he was "extremely concerned" about fuel supplies not being dispersed across the country. "I'm even more concerned when I get a message this morning from my colleague, Minister James Brown, about curtailment of fire and emergency services," he told Newstalk on Friday. "No-one can stand over that. We do not want to have to deal with this situation in a heavy-handed way and I don't think that will happen. What we need to do is to de-escalate this now, today." Enterprise Minister Peter Burke said that the Government was working on a package of measures to combat inflation, which he said could be announced on Friday. "I do believe we will have news on that, in terms of an intervention that will reduce inflation, that will try and reduce the price of groceries on the shelves for our consumers, for vulnerable people buying them," he said. Meanwhile, Government ministers are meeting with representative bodies of farmers and hauliers to discuss further support to deal with the high cost of fuel on Friday afternoon. Protesters, including one of their spokesmen Mr Dallon, were refused entry to the meeting at the Department of Agriculture. Mr Dallon said protesters had been told by TDs he would be permitted to attend the meeting but that he was turned away when he arrived. He said "what we really wanted was to be able to step down the protest" and attend the meeting to "get results", adding "we had demands". Asked if the protests would continue, Mr Dallon said: "They'll continue until we get a result." Ger Hyland, president of the Irish Road Haulage Association (IRHA), has offered to act as an "honest broker" between protesters and the Government. He said he had been in contact with the protesters "through intermediaries" and is available to meet some of them before the meeting and take their concerns to the Government.
European airports warn of half-term holiday chaos due to jet fuel shortage
Add Yahoo as a preferred source to see more of our stories on Google. Airports across Europe could face "systemic" jet fuel shortages within three weeks if the Strait of Hormuz remains closed, an airport industry group has warned. Airports Council International Europe, which represents more than 600 airports in 55 countries, sent a letter to the European Commission, warning that a fuel crunch would "significantly harm the European economy". The group said "the impact of military activity on demand" was further straining supplies. With peak summer season nearing, the fuel shortages have prompted fears of delays and cancellations as millions of Britons prepare to travel for the May half-term school break. In the letter seen by the Financial Times, EU transport commissioner Apostolos Tzitzikostas was warned of "increasing concerns of the airport industry over the availability of jet fuel as well as the need for proactive EU monitoring and action". "If the passage through the Strait of Hormuz does not resume in any significant and stable way within the next three weeks, systemic jet fuel shortage is set to become a reality for the EU." Jet fuel prices have more than doubled to $150 to $200 per barrel in recent weeks, as the key global shipping route, which carries much of the world's jet fuel supply, has stayed closed. The airport body said its concerns had been intensified by the approach of the holiday season, "when air travel enables the whole tourism ecosystem upon which many [EU] economies rely". According to a study from the group, air connectivity contributes €851 billion (£741 billion) in gross domestic product for European economies each year and supports 14 million jobs.
SpaceX IPO Targets $1.75T Valuation, but History Signals Risk of Early Stock Dip
SpaceX, owned by Elon Musk, has filed for an IPO. Reports say the company is aiming for a valuation of $1.75 trillion. It also plans to raise about $75 billion. Interest remains high due to the rapid growth in space and the AI boom. Experts say the space sector could reach $1 trillion in the next decade. Artificial intelligence is also expanding quickly and could add huge value to the global economy. These trends have increased hopes around SpaceX. However, IPO history tells a different story. Many companies failed to perform well after going public. Names like Group, Meta Platforms, and Saudi Aramco saw their shares drop within months. Only Visa showed strong early gains.
Palantir stock rebounds after Trump endorsement amid Anthropic competition By Investing.com
Investing.com -- Palantir Technologies (NASDAQ:PLTR) shares reversed off session lows Friday after President Trump mentioned the company in a social media post, though the stock remained down 2% after falling as much as 6% earlier in the session. Get premium commodity market insights with InvestingPro subscription Trump wrote, "Palantir Technologies (PLTR) has proven to have great war fighting capabilities and equipment. Just ask our enemies!!!" The endorsement came as the stock faced selling pressure amid concerns about competitive threats from Anthropic. Palantir shares had declined over 15% in the past five days before Trump's mention today. The recent weakness followed Anthropic's release of a new product around multi-agent orchestration, which raised questions about potential headwinds for the software sector. Wedbush analyst Dan Ives earlier reiterated an Outperform rating and $230.00 price target on Palantir, addressing the competitive concerns. The analyst commented, "Palantir has been under pressure over the past few days, including down 7% today, after Anthropic released a new product around multi-agent orchestration, which continues to add more headwinds to the software sector. While Anthropic is hitting a new scale with the company now at $30 billion ARR, up from $9 billion at the start of the year, we believe this is not at the expense of PLTR's business as the company continues to accelerate both its US commercial and government businesses with US Commercial growing 137% YoY and US government accelerating 66% YoY." Ives noted that Palantir's US commercial business grew 137% YoY while US government revenue accelerated 66% YoY, suggesting the company's growth trajectory remains intact despite competitive developments in the AI sector.
PMI Agile Alliance unveils the "Manifesto for Enterprise Agility", a bold, new guide to power enterprise-wide reinvention
PMI Agile Alliance has released the Manifesto for Enterprise Agility, a leadership guide for organisations facing frequent disruption and rising pressure to reinvent. PMI global C-suite research shows that reinvention is the norm: 93% of senior executives say they must rethink and challenge assumptions of their operating models or business approaches at least every five years, and nearly 65% say they are doing so every two years or faster. The challenge is not recognising change and the need to adapt faster; it's converting strategy into action. That strategy-execution gap is where enterprise agility becomes essential - but where ambition still outpaces reality. While 85% of C-suite executives recognise enterprise agility as critical and very important, 65% admit they implemented it to a limited extent or not at all. "Most organisations don't struggle with strategy; they struggle with turning strategy into coordinated action. Enterprise agility is about building organisations that can adapt quickly without losing alignment, so leaders can respond to disruption while keeping their people and priorities focused on delivering value," says George Asamani, MD, Project Management Institute, Sub-Saharan Africa. Launched in the 25th anniversary year of the Manifesto for Agile Software Development, the Manifesto for Enterprise Agility moves agility beyond teams and projects to the entire enterprise -- including leadership behaviour, operating models, governance, and culture. Rather than prescribing a framework, the Manifesto focuses on how leaders build and run the system for enterprise-level agility - governing with guardrails instead of gatekeepers, funding intent instead of activity, and moving authority closer to where value is created. The Manifesto for Enterprise Agility is for organisations that need to adapt faster, stay aligned, and keep strategy actionable. The principles guide executives and practitioners in operationalising the values and offer leaders the clarity to act on what really matters. Endorsers of the Manifesto describe why that matters now: Greg Beato, co-author of Superagency, "Twenty-five years after the Manifesto for Agile Software Development presented a new way to think about software development, it's time to apply similar thinking to enterprises as a whole, not just to projects or products. Just as the Agile Manifesto was a response to a major change in technological conditions driven by the internet, the growth in both physical and digital networks around the world compels enterprises to incorporate and deploy agility to their entire organisational systems, including leadership, operating models, execution governance, and culture." Kevin Nolan, CEO of GE Appliances, "Today's business landscape demands rapid adaptation and greater agility. Agile organisations adapt faster and take the lead, while those not embracing agility risk falling behind as collaboration becomes essential in a dynamic environment." Sagar Kochhar, former CEO and co-founder of Rebel Foods, "Enterprise agility is less about frameworks and more about leadership courage - the courage to reset the vision, dismantle legacy assumptions, and trust teams to execute within systems designed for speed. This Manifesto captures a critical truth: enterprise agility is not a transformation initiative, but a leadership mindset required to continuously reinvent vision, structure, and execution in a volatile world." The Manifesto is grounded in PMI® research, including global C-suite surveys, executive interviews, and input from senior transformation practitioners, reflecting the realities leaders face across industries.
Anthropic's Mythos is changing everything we knew about AI - Alarm over cyberattacks, what banks and Wall Street fear - ProtoThema English
The software managed to identify vulnerabilities that no one had found for decades in systems known for their security - An unprecedented risk of cyberattacks if it falls into the wrong hands Alarm has been raised within the U.S. government over Anthropic's new artificial intelligence model, "Claude Mythos Preview," with Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell convening an emergency meeting with heads of major banks in Washington to assess potential cyber risks and urge financial institutions to reinforce their systems in time. The trigger for this emergency response was Anthropic's own admission that Mythos is so powerful that it cannot yet be released to the general public. The company announced that it has suspended its general rollout and is, for now, limiting it to a closed collaboration framework with major technology and financial companies, until it determines what safeguards and security rules are required before any wider release. This move is telling in itself: one of the leading companies in the sector is publicly stating that its new model is so powerful that normal commercial release would be premature and potentially dangerous. In practice, Mythos has been included in the new "Project Glasswing," an Anthropic initiative involving companies such as Amazon Web Services, Apple, Microsoft, Google, Cisco, Nvidia, Palo Alto Networks, CrowdStrike and JPMorgan Chase. The rationale behind the initiative is that organizations managing critical digital infrastructure will gain first access to the model, allowing them to test their own systems, identify vulnerabilities and share findings before such a tool reaches hands that could use it offensively. The concern is not theoretical. Anthropic has reported that, in internal testing, Mythos was able to identify and then exploit weaknesses "in every major operating system and every major web browser" when given the appropriate instruction by a user. In some cases, the model reportedly discovered thousands of serious vulnerabilities, including long-standing flaws that had evaded conventional auditing tools for years. Close It found decades-old vulnerabilities that no one had detected Anthropic does not describe Mythos in vague terms. In its technical documentation, it states that in internal testing the model was able to identify and exploit zero-day vulnerabilities "in every major operating system and every major web browser," when prompted by a user. Moreover, it did not rely only on simple or known techniques: in one case it generated a browser exploit that combined four different vulnerabilities, bypassing protection layers of both the browser and the operating system. In another, it identified a 27-year-old bug in OpenBSD, an operating system known for its strong security reputation, and also found a 16-year-old vulnerability in FFmpeg, a library that has been extensively audited for years using automated tools. What banks really fear This is precisely where the deeper concern of Washington and the financial system lies. Large banks and markets do not rely only on "modern" cloud systems, but also on vast layers of legacy software, browsers, endpoint environments, middleware, supply chains and applications connecting payments, trading, custody, data feeds and internal communication. A system like Mythos does not simply mean that a new vulnerability can be found. It means that it dramatically shortens the time between the discovery of a weakness and its operational exploitation. In simple terms, it reduces the window in which defenders can fix a problem before it turns into a real attack. This is the dimension that explains why the issue reached directly the Treasury and the Fed. Anthropic argues that Mythos was not originally designed as a cybersecurity tool, but as a general-purpose model with highly advanced capabilities in reasoning, programming and solving complex problems. However, it is precisely these capabilities that, according to the company, allow it to function as an extremely effective mechanism for identifying -- or even exploiting -- security gaps. For this reason, the company has already discussed the model's "offensive and defensive" cyber capabilities with U.S. authorities, without specifying which agencies were involved. The message now coming from Washington is that the issue goes far beyond a typical corporate announcement of a new AI product. The emergency meeting at the U.S. Treasury, with the participation of Bessent and Powell themselves, shows that American financial authorities see in Mythos a preview of the next major threat: AI models so advanced that they can function simultaneously as a defensive tool and as a weapon for large-scale cyberattacks. Anthropic insists that it is trying to give "an advantage to the defenders." For this reason, Glasswing is accompanied by $100 million in usage credits for participants, as well as additional funding for open-source and security organizations, while the company has committed to publishing a public report within 90 days on findings, vulnerabilities fixed and conclusions that can be shared without creating new risks. The goal is to shape new standards for vulnerability disclosure, patching, software supply chains, secure-by-design practices and frameworks for regulated sectors. The case also has a second reading for the market. Anthropic, already at the center of the global race for ever more powerful AI models, is now trying to set a precedent: access first to "defenders," then to the public. Whether this logic will become the new standard for the most advanced AI systems remains to be seen. For now, however, Washington and Wall Street are treating Mythos as a technological leap that could change the rules of the game in cybersecurity. Meta joins the race for "superintelligence" with Muse Spark Meta is also stepping up in the AI race, presenting its new model Muse Spark, the result of a costly investment aimed at developing "superintelligence," meaning AI that can surpass human cognitive abilities. Muse Spark is the first model in a new series, internally known as Avocado. Unlike previous Llama models, Meta is not proceeding with a fully open release, but is initially offering only limited access ("private preview") to selected partners. The model will first be available via the Meta AI app and will gradually be integrated into platforms such as WhatsApp, Instagram, Facebook and the company's smart glasses. According to Meta, Muse Spark has been designed to be "small and fast," yet capable of handling complex queries in fields such as science, mathematics and health. Independent evaluations show that the model approaches the performance of top competitors in language and visual understanding but lags in programming and abstract reasoning. It ranked fourth overall in the Artificial Analysis benchmark. The company's CEO, Mark Zuckerberg, has already hinted at gradual performance improvements, noting that the first models will mainly demonstrate Meta's "rapid trajectory of development."
Fuel Protests Cause Transport Chaos in Ireland as Iran War Spikes Prices
The Irish government said it had called in the army to help clear blockades of crucial roads, after days of protests over the surging price of fuel, driven by the war in the Middle East, brought highways and streets to a standstill. The government said in a statement that it had held an emergency meeting on Thursday night to discuss the effects of the protests, which it said were causing "significant disruption for the public, to supply chains and vital services." It added that the blockades had restricted access to a number of ports, preventing fuel from being distributed to service stations in some parts of Ireland and causing concern that emergency vehicles could run out of fuel. Groups of protesters began blocking roads on Tuesday with trucks and tractors, demanding the government do more to ease the burden of fuel costs, which have jumped in price around the world since the start of the U.S.-Israeli war with Iran. The protests appear to have been organized largely in messaging groups and on social media, with truckers, farmers and others reliant on fuel for their businesses taking part. Last month, the Irish government cut taxes on gasoline and diesel fuel and suspended a tax on home heating oil. But many protesters said the actions did not go far enough, with some demanding a cap on the cost of diesel and gasoline, while others want the tax cuts to be extended. On Friday, trucks and buses blocked O'Connell Street Bridge in Dublin, a main artery connecting the north and south of the capital, while other protesters blocked roads connecting key ports and the country's only oil refinery in Cork. Micheál Martin, the Irish prime minister, said in an interview with The Irish Times published on Friday morning that the government was open to engaging with the protesters but only "through established negotiating channels." He said that further measures to ease costs could be possible through negotiations, but added, "the strikes will have to end first." The Irish government met with a number of unions this week, including on Monday, the day before the protests began. Representatives from a variety of unions, including the Irish Farmers' Association, are set to meet with the government on Friday afternoon. Some of the most disruptive actions have taken place on the M50, the highway that encircles Dublin. It is also the main route to the country's busiest flight hub, Dublin Airport. Some travelers stuck in traffic abandoned their taxis and buses and walked to the airport, rolling their bags along the shoulder, in an effort to make it to their flights after dozens of trucks parked on the roadway on Thursday evening. But the protests were not just focused on the capital, with demonstrations in towns and villages across the country also causing standstills on local roads. In the small village of Inch in County Wexford, dozens of cars lined an overpass, with protesters waving an Irish flag and cheering. Tractors blocked the on ramp to the highway, forcing frustrated motorists to turn back and find alternative routes.
Kraken Reveals Fed Account Bars Interest and Emergency Lending | PYMNTS.com
By completing this form, you agree to receive marketing communications from PYMNTS and to the sharing of your information with our sponsor, if applicable, in accordance with our Privacy Policy and Terms and Conditions. Kraken can't earn interest on reserve balances held at the Fed, can't access emergency Fed lending, and can't access the Fed's FedNow® Service and ACH payments systems, according to the report, which cited a Kraken spokesperson. These details were not previously disclosed, according to the report. The company's Fed master account does allow its Wyoming banking arm to access the Fedwire wholesale payments system and hold limited balances overnight, per the report. "We look at this as a great testament to regulatory rigor and cooperation," Jonathan Jachym, global head of policy at Kraken, told Reuters. "It promotes principles of both safety and soundness, and innovation." PYMNTS reported March 4 that the approval granting Kraken Financial a Federal Reserve master account followed years of "sustained regulatory engagement" and made Kraken Financial the first bank of its kind to be given access to the Fed's payment system. According to Friday's report by Reuters, the Kansas City Fed's decision to grant Kraken a "limited purpose account" last month led to banks raising concerns about risk, and Rep. Maxine Waters, the ranking member of the House Financial Services Committee, questioning the transparency of the decision. Waters asked the Kansas City Fed in March to disclose more details about the decision by Friday. A Kansas City Fed spokesperson told Reuters that it is reviewing the letter. Waters said in a March 26 press release about the letter: "The Kansas City Fed's announcement does not disclose specific information about Kraken's access to the range of Federal Reserve financial services 'due to the confidentiality of business information provided by applicants.' However, the announcement raises questions about the approval because neither statute nor the Federal Reserve Board's Account Access Guidelines refer to a 'limited purpose account' type." When announcing the approval of a limited purpose account for Kraken in a March 4 press release, Kansas City Fed President Jeff Schmid said: "As we know, the payments landscape is actively evolving. Throughout this transformation, the integrity and stability of the U.S. payments system remain our priority." The Kraken spokesperson told Reuters, per the report, that the company's bank reserves are fully backed, the company complies with all bank-grade anti-money laundering (AML) and know your customer (KYC) requirements, and the company has never been hacked.
AI that finds and exploits bugs? Key facts about Anthropic's new AI model Mythos raising red flags
Anthropic AI Claude Mythos Preview explained: A new artificial intelligence model from Anthropic is drawing serious attention from experts and engineers, with some warning it could reshape cybersecurity and potentially make hacking more advanced than ever. The model, called Claude Mythos Preview, has not been released publicly. Instead, it has been given to a limited group of tech companies due to concerns about the risks it could pose if widely available. Anthropic researchers say Mythos Preview was able to detect thousands of high- and critical-severity bugs across major operating systems and web browsers, as per a NBC News report. Some of these vulnerabilities had reportedly gone undiscovered for decades. What's raising concern is not just its ability to find flaws, but also to act on them. According to Anthropic, the model can identify vulnerabilities, write code to exploit them, and combine multiple weaknesses to break into complex systems, all on its own. Katie Moussouris, CEO of Luta Security, said the concerns surrounding the model are not exaggerated, warning of "huge ramifications" as such technology develops. At the same time, some researchers have urged caution, noting that limited public details make it difficult to fully verify the claims. Rather than a full release, Anthropic launched an initiative called Project Glasswing, giving over 50 organizations, including companies like Microsoft, Nvidia, and Cisco, access to the model. The goal is to help strengthen cyber defenses before such capabilities become widespread. The company has also said it will disclose details about the vulnerabilities within 135 days of informing affected organizations, as per the NBC News report. This is the first time in years that a major AI company has chosen to restrict a model's release over safety concerns. The decision echoes OpenAI's move in 2019 to limit access to its GPT-2 system due to potential misuse. Anthropic's cyber researcher Logan Graham said, "We are not confident that everybody should have access right now," adding, "We need to start figuring out how we'd prepare for a world of this first before we can handle the idea of black hat [criminal or adversarial] hackers having access," as quoted by NBC News. Anthropic has briefed US government officials on Mythos Preview's capabilities, including both its defensive and offensive potential. The company is also involved in an ongoing dispute with the Trump administration after being labeled a "supply chain risk to national security," a designation currently under legal challenge. Beyond cybersecurity risks, internal testing revealed unusual behavior. In one evaluation, the model appeared aware it was being tested in about 29% of cases, as per the NBC News report. In another, it seemed to intentionally perform worse to avoid drawing attention, something researchers say they had not seen before. In a separate experiment, an earlier version of the model even managed to send a message outside a restricted system, despite not being expected to have internet access. What is Claude Mythos Preview? It is a new AI model by Anthropic with advanced cybersecurity capabilities. Why hasn't the model been released publicly? Anthropic is limiting access due to concerns about potential misuse.
CoreWeave (CRWV) Jumps on New Anthropic Deal - Wall Street Pit
* CoreWeave (CRWV) has signed a multi-year agreement to supply Anthropic with cloud computing capacity for its Claude AI models, with initial rollout beginning later this year and potential for future expansion. * The announcement drove CoreWeave shares up more than 9% to $100.02 in early Friday trading, building on a series of major recent deals including $11.9 billion with OpenAI, $6.3 billion with Nvidia (NVDA), and $21 billion with Meta (META). * The partnership further diversifies CoreWeave's revenue base, which was previously dominated by Microsoft (MSFT) at about 67% last year, and strengthens its position as a leading specialized provider of high-performance AI infrastructure. CoreWeave (CRWV) has secured a multi-year agreement to supply Anthropic with significant cloud computing capacity, set to come online later this year and support workloads for the Claude family of AI models. The deal features a phased infrastructure rollout with potential for future expansion, further strengthening CoreWeave's position as a specialized provider of high-performance AI infrastructure. The announcement drove CoreWeave shares up 9% to $100.02 in early Friday trading. This latest partnership adds to a rapid series of major contracts, including an $11.9 billion agreement with OpenAI last year, a $6.3 billion initial order with Nvidia (NVDA) in September, and an expanded $21 billion deal with Meta (META) announced on Thursday. These agreements are helping CoreWeave diversify its revenue base, which last year was heavily concentrated with Microsoft (MSFT) accounting for about 67% of total revenue. As a neocloud operator with deep ties to Nvidia, CoreWeave has emerged as a critical intermediary in the AI supply chain, delivering the dense, high-performance clusters that frontier AI labs require. The Anthropic deal underscores growing demand for flexible, purpose-built compute capacity at a time when leading AI developers are racing to scale training and inference operations while reducing reliance on traditional hyperscalers. With Anthropic itself recently expanding capacity through agreements with Broadcom and Google and exploring its own chip designs, CoreWeave's ability to deliver specialized infrastructure positions it as an increasingly important player in the evolving AI ecosystem. WallStreetPit does not provide investment advice. All rights reserved.
Southaven residents and advocate groups appeal state's decision to grant xAI permit | DeSoto County News
A cohort of residents from Southaven, as well as two nonprofit advocacy groups, are petitioning the Mississippi Department of Environmental Quality to hold a hearing with hopes of reversing the state's approval of an air permit for data center company xAI. The Southern Environmental Law Center sent MDEQ the appeal Thursday on behalf of the Safe and Sound Coalition -- a group of Southaven residents pushing back against xAI's expansion in the city -- the NAACP and Young, Gifted & Green. The letter to the state comes about a month after the Mississippi permit board approved an air permit for 41 new, permanent gas turbines at an xAI facility in Southaven. The turbines would be part of a network of generators powering xAI's data center operation in the area, which includes two centers in Memphis as well as a recently announced facility set to open in Southaven. After frustrating experiences with noise and unchecked emissions from xAI's temporary turbines in the community, dozens of Southaven residents have spoken out against the company's operation, both at a February town hall and at the March permit board hearing. "(The permit board's) rushed approval of this flawed air permit raises serious concerns about transparency, regulatory accountability, and whether the public is truly being heard in decisions that directly affect our health and environment," said Shannon Samsa with the Safe and Sound Coalition. "Those with the authority to protect us had every opportunity to slow this process down, fully evaluate the risks, and ensure meaningful public involvement. Instead, they chose to push it forward." MDEQ, which presented the permit to the board for consideration, said its air modeling suggests the new turbines would keep the area in compliance with national air quality standards. The Southern Environmental Law Center, though, argues the generators would release "staggering amounts of air pollution and harmful chemicals like formaldehyde" in "an area failing to meet EPA's ambient air quality standards." The Thursday letter to MDEQ seeks an evidentiary hearing to discuss "deficiencies" with the permit granted to xAI, a company owned by billionaire Elon Musk that uses its Memphis data centers to power the AI chatbot Grok. "This permit is unable to clear the most basic hurdle by failing to show that pollution from xAI's 41 turbines won't hurt air quality," law center attorney Patrick Anderson said. "This appeal aims to fix these glaring errors while holding the agency accountable."
Anthropic's Claude Mythos vs Meta's Muse Spark: Everything You Need to Know.
Two very different AI companies. Two closed models. One wild week. Something strange happened this week. Meta: the company that built its entire AI reputation on open-source Llama models. Anthropic: the company that accidentally leaked Claude Code's source via an npm packaging error last month. If you can't read the article further due to paywall than please click here Both dropped new models within 48 hours of each other. Both models are closed. But only one of them was ever expected to be open. But the reasons could not be more different. What Actually Happened * April 7: Anthropic announced Claude Mythos Preview alongside Project Glasswing. The model is not publicly available. It is accessible via the Claude API, Amazon Bedrock, Google Vertex AI, and Microsoft Foundry, but only for invited Glasswing partners. Anthropic restricted access because the model found thousands of zero-day vulnerabilities across every major operating system and browser. Autonomously. * April 8: Meta announced Muse Spark, the first model from Meta Superintelligence Labs. It powers the Meta AI app and website today. No open weights. No public API. A private API preview exists for "select partners." This from the company that averaged one million Llama downloads per day. Two closed models. Two very different stories. The Head-to-Head Comparison Here is how these two models stack up across the dimensions that matter: Table 1: Overview and Access Table 2: Shared Benchmarks (Head-to-Head) These are the benchmarks where both models have reported scores. Direct comparison.
UbiQTree: Uncertainty quantification in XAI with tree ensembles
Machine learning (ML) is a key part of improving healthcare analytics such as resource planning, disease diagnosis, prognosis, and risk stratification. However powerful they are, uncertainty is inherent and ubiquitous in ML models because their predictions are affected by noisy data, model limitations, and unseen scenarios. To address this challenge, some of the most widely used tools are ensemble tree-based models, which help in managing and quantifying uncertainty in predictions. They are highly accurate, interpretable, and efficient with structured data, resulting in lower computational demand. Unlike deep neural networks, which require large amounts of unstructured data such as images and text, they have lower computational requirements and are more interpretable. These include random forest (RF), gradient boosting machines (GBM), and extreme gradient boosting (XGBoost). These models are robust against noise and can handle large, complicated datasets, which are common in healthcare. Ensemble tree approaches are different from traditional ML models as they can efficiently capture complex, nonlinear relationships and slight interactions among the features. This leads to highly accurate and generalizable predictions. Ensemble models have many advantages. They combine the strengths of multiple base learners which reduces overfitting, improves stability, and enhances the model's ability to generalize to unseen data unlike traditional ML models. It has been shown that using a group of classifiers to make predictions outperforms using individual classifiers to predict heart disease. This is important for making reliable decisions, which makes them very useful in healthcare. Despite their strengths, ensemble tree-based models have two main problems. First, they are difficult to interpret, particularly when there are large numbers of constituent trees and features. In model selection, the first choice is to avoid this "black box" nature. To address this, explainable AI (XAI) techniques such as Shapley additive explanations (SHAP) or Shapley values, which are rooted in cooperative game theory, have emerged as principled frameworks for attributing the contribution of each feature to individual predictions in ML models. The field of XAI is rapidly evolving: counterfactual paths (CPATH) identifies feature permutations that influence model predictions and uses domain knowledge graphs; total causal effect calculation for fuzzy cognitive maps (TCEC-FCM) uses graph traversal techniques to compute causal effects, improving system transparency; and reciprocal human-machine learning (RHML) fosters continuous learning between humans and AI models, improving model performance and decision-making. Other recent methods include diverse counterfactual explanations (DiCE), which provides counterfactual explanations that align with human cognitive processes and are highly valued for understanding a model. Logic tensor networks (LTN) is a framework rooted in neural-symbolic AI, designed for learning and logical reasoning. It facilitates interactive explainability and model revision when applied to XAI. Template system for natural language explanations (TS4NLE) is designed for the presentation and rendering of explanations derived from other XAI outputs, prioritizing human comprehensibility. It can process the structured output of any approach and generate natural language explanations (NLEs), which are then rendered into tailored natural language text using a template system. These frameworks aim to make AI systems more transparent and aligned with human understanding. Despite these advancements, SHAP remains a popular candidate for studying XAI due to its strong theoretical foundation and widespread adoption. However, calculating Shapley values can be difficult for complex models. Fortunately, a couple of new, efficient methods for calculating them for certain types of models have recently emerged such as TreeSHAP. TreeSHAP calculates Shapley values in tree-based models including decision trees, RFs, and XGBoost in a fast and efficient way. TreeSHAP uses the natural structure of decision trees to make predictions that are much faster and easier to understand than those of other methods. This is helpful for explaining results with large groups of data, which is important in fields where understanding predictions is paramount, for instance in healthcare or finance. SHAP values provide a clear framework for determining how each feature contributes to individual predictions. They offer insight into the decision-making process behind complex ensemble models. SHAP and similar methods not only encourage trust but also make it easier to use advanced ML models in healthcare by making them easier to comprehend.
SpaceX plans Vandenberg rocket launch today. How to watch live stream
Less than an hour before the Artemis II astronauts are due to land near California, SpaceX could launch a rocket about 200 miles north. Hours after the Artemis II astronauts are due to splash down off the coast of San Diego, California, SpaceX could launch a rocket about 200 miles north. A Falcon 9 rocket is due to get off the ground Friday, April 10 from the Vandenberg Space Force Base in Santa Barbara County. Its mission? Deploy SpaceX's Starlink broadband internet satellites into low-Earth orbit. The launch window is due to open shortly after the four Artemis II astronauts are due to return from a 10-day mission around the moon and make a water landing in the Pacific Ocean. And both will have a livestream available for those who want to watch. Here's everything to know about the latest SpaceX mission, and how to watch a webcast of the Falcon 9 launching in Santa Barbara County. Is there a rocket launch today? Next liftoff from California SpaceX is working toward a Friday, April 10, launch from Southern California, with a four-hour launch window opening at 7:39 p.m. PT, according to a launch alert. The launch will take place from the Vandenberg Space Force Base in Santa Barbara County. A Federal Aviation Administration operations plan advisory suggests a backup opportunity is available the next day if the launch were to be postponed. What is launching from Vandenberg? Falcon 9 to deploy Starlink satellites SpaceX will launch its famous two-stage 230-foot Falcon 9 rocket, one of the world's most active, to deliver 25 Starlink satellites into low-Earth orbit, an altitude nearer Earth's atmosphere where they're able to circle the planet quickly. How to watch SpaceX launch livestream Californians, of course, have plenty of opportunities to see a rocket in person both near the launch site as it lifts off, and further away as it soars overhead. But SpaceX also provides a live webcast of its missions for those who prefer to watch from home or for those viewing the launch locally and looking for updates in real-time. As with most SpaceX missions, the launch will be available to stream on the company's website and its new X TV mobile app, beginning about five minutes before liftoff. SpaceX may also provide updates on social media site X. Does Elon Musk own SpaceX? What to know about rocket company SpaceX is the commercial spaceflight company that billionaire Elon Musk, the world's richest man, founded in 2002 and leads as the CEO. SpaceX is headquartered at Starbase in South Texas near the U.S.-Mexico border. The site, which is where SpaceX has been conducting routine flight tests of its 400-foot megarocket known as Starship, was recently voted by residents to become its own city. As a major government contractor, SpaceX serves as the launch service provider for a variety of government missions both civil and military. For the Department of Defense, SpaceX's Falcon 9 helps launch classified satellites and other payloads into space. And for NASA, Falcon 9 most often helps propel astronauts to the International Space Station on SpaceX's Dragon crew capsule - the only U.S vehicle capable of carrying NASA astronauts to orbit. What is Starlink? Starlink is SpaceX's internet satellite business. With more than 10,000 satellites in its growing orbital constellation, Starlink has become a lucrative part of Musk's business empire, serving millions of customers around the world. Eric Lagatta is the Space Connect reporter for the USA TODAY Network. Reach him at [email protected]
UniPat AI Launches EchoZ Prediction Model, Demonstrating Performance Beyond Human Traders on Polymarket
BEIJING, April 10, 2026 (GLOBE NEWSWIRE) -- UniPat AI announces benchmark results for its forecasting model, EchoZ-1. 0 after testing in sandbox and live environments on Polymarket. Prediction markets are a place where individuals can trade on the results of a future event. Third-party data, including from Dune Analytics, show diverse participant performance over time in such liquid (and competitive) markets. According to UniPat AI, EchoZ-1. 0 achieved a 63.2 % alignment rate of its outputs on questions about politics and governance during testing. For predictions based on a seven-day period or longer, the model predicted alignment 59.3 % of the time. In high-uncertainty cases that the company categorized as such, with baseline confidence scores "between 55 to 70 per cent," their reported alignment rate was 57.9 per cent. The companies said these results were obtained under specific test conditions and do not necessarily predict future performance. Five different agents were deployed autonomously to benchmark performance in real-world scenarios, all driven by the EchoZ-1. 0 model for a week. Overall, four agents produced positive returns during that time period, and one did not. These results were subject to market conditions and execution factors, UniPat AI said. The model was tested against scenarios with multiple variables, the company said, which included regulatory developments, geopolitical occurrences, on-chain governance decisions and market-related events. These scenarios were chosen to represent situations with partial or evolving information. UniPat AI has also performed a set of stress tests regarding its evaluation framework. These modifications consisted of changing scoring parameters, partially removing input data, and modifying model configurations. Under these conditions, models maintained consistent rankings within the company's testing framework, it said. UniPat AI claims it has released prediction inputs, probability estimates, timestamps and ultimate outcomes to the public for independent auditing. It consists of aligned and non-aligned predictions. Examples it cites include forecasts for equity market capitalisation, digital asset price levels and professional sports standings. Probability estimates were documented before outcome resolution in each case for validation. EchoZ-1. 0 relies on a method named "Train-on-Future," UniPat AI said, and differs from methods that mainly use historical data. It includes evaluation criteria to capture reasoning structure as well as outcome alignment. In this process, a system called Automated Rubric Search is used to extract reasoning patterns related to the observed actions. This development team consists of researchers with expertise in reinforcement learning, data synthesis, and model evaluation. The prediction markets were chosen as a testing ground because of their measurable and verifiable nature, UniPat AI said. A spokesman for the company said structured assessments of probabilities may have applications in places where there's decision-making under uncertainty. About UniPat AI UniPat AI is a research-focused artificial intelligence company working on the development of machine learning systems designed for real-world applications. The organisation describes its mission as advancing AI systems from experimental models to practical, deployable tools capable of operating in complex environments. Media Contact Organization: UniPat AI Contact Person Name: Yao He Website: https://unipat.ai/ Email: [email protected] Disclaimer: This sponsored content reflects the views of the content provider only and not those of this media platform or its publisher. It is for informational purposes and not financial, investment, or business advice. All investments carry risks, including loss of capital. Readers should do their own research and consult a qualified advisor before making decisions. Speculate only with funds that you can afford to lose.The media platform and publisher are not responsible for any inaccuracies or losses. GlobeNewswire does not endorse any content on this page. Legal Disclaimer: This article is provided on an "as-is" basis, without warranties or representations of any kind, express or implied. The media platform assumes no responsibility or liability for the accuracy, content, completeness, legality, or reliability of the information presented. Any complaints, claims, or copyright concerns related to this article should be directed to the content provider mentioned above. A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/6db3c54f-65e4-408f-b58b-f530a3cfffab
Fed's Powell, Scott Bessent warn bank CEOs of Anthropic AI risk: BBG
Anthropic's (ANTH.PVT) new Claude Mythos have raised alarm bells this week as the AI model sparks concerns from its own developer that it could enable hackers. Morning Brief Host Julie Hyman and Yahoo Finance Head of News Myles Udland react to Bloomberg reporting that US Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell met with banking CEOs to warn them of the possible cyber threat to financial institutions that this Claude version presents.
Fuel-price protesters cause chaos in Ireland and spread to Norway
Add Yahoo as a preferred source to see more of our stories on Google. Protests over fuel prices have caused chaos in Ireland and spread to Norway in a knock-on effect from the conflict in the Middle East. Hauliers, farmers and other groups blocked motorways and brought parts of Dublin to a standstill on Friday in a fourth consecutive day of action. In Ireland the protests have sparked fuel shortages and travel disruption, and in Norway lorry drivers taking part in the "diesel roar" protest descended on the capital. The Irish government put the army on standby to help remove blockades and police warned some protesters to disperse or face arrest, prompting defiance and threats to continue the disruption for weeks if necessary. Protests were endangering critical supplies of food, fuel, clean water and animal feed, the police force, An Garda Síochána, said in a statement. "This is not tolerable and is against the law." Government leaders have accused protesters of holding the country to "ransom". The blockade of ports and a refinery meant Ireland was on the verge of turning away oil deliveries and losing its supply, the taoiseach, Mícheál Martin, told RTÉ. "It is unconscionable, it's illogical." Despite government mitigation measures, in recent weeks the price of diesel has risen from about €1.70 a litre to €2.17 and petrol has jumped from about €1.74 to €1.97. Industry representatives were expected to convey their members' grievances at a meeting with ministers later on Friday but it was unclear whether that would satisfy protesters who have called for direct talks with the government. The justice minister, Jim O'Callaghan, said "outside actors", such as the British far-right activist Tommy Robinson, were manipulating the protests for their own agenda. In Denmark's recent election the far-right Danish People's Party tried to tap discontent by paying voters for their petrol. The rise in oil prices since the US and Israel began attacking Iran on 28 February has convulsed global markets and triggered outcries from consumers and businesses who want governments to do more to soften the blow. Some announced temporary cuts in fuel taxes while others took measures to restrict demand and considered rationing. The Philippines declared a state of "national energy emergency". Authorities in France tried to avert widespread shortages by announcing on Friday that fuel tankers would be allowed to circulate on weekends and public holidays until 11 May. In Norway protesters on Friday drove a convoy of lorries to the parliament in Oslo. About 70 to 80 trucks, some with banners that read "nok er nok!" (enough is enough!), joined another group known as Dieselbrølet (diesel roar). Only a handful were allowed to drive into the capital. Norway cut fuel taxes on 1 April but hauliers say they need more predictable and lower prices. Despite being an oil producer, fuel prices in Norway have surged since the effective closure of the strait of Hormuz. The Statistics Norway institute said the price of fuel and lubricants rose by 17.9% from February to March, with diesel prices in that period jumping by 23.6%. A Statistics Norway spokesperson said it had never recorded a sharper month-on-month increase in fuel prices using the CPI inflation index. "The last time we saw something similar was in the spring of 2022, following Russia's full-scale invasion of Ukraine, but in that case the price increase occurred over two consecutive months." Last month the Irish government announced a €250m package of measures to reduce fuel costs, including a temporary excise duty reduction, expansion of a diesel rebate scheme for hauliers and bus operators and an extension of the fuel allowance. Blockades of Ireland's sole oil refinery at Whitegate, Co Cork, and fuel depots in Galway City and Foynes in Co Limerick crippled deliveries. Dozens of forecourts ran dry and there were warnings that would soon become hundreds as motorists rushed to fill up on petrol and diesel. Columns of tractors and other vehicles closed motorways and Dublin's main thoroughfare, O'Connell street. The Irish Medical Organisation said slower emergency services response times and missed healthcare appointments would harm patient welfare. The courier company DPD suspended deliveries. Protesters were prepared to remain in the capital for weeks, a spokesperson, John Dallon, told RTÉ. "If it takes a month, we are prepared to sit here," he said. He accused the government of ignoring the plight of people facing hardship and ruin because of fuel costs. "How dare they come out and say that these people that are protesting are holding the country to ransom? It's the government that's holding this country to ransom, not the protesters." The taoiseach postponed a trade mission to Canada to deal with the crisis.