News & Updates
The latest news and updates from companies in the WLTH portfolio.
Nobody Wants Anthropic's Best AI Model Anymore Now That There Are Way Cheaper Alternatives
One thing that could put a dent in the hype around Anthropic's upcoming, multitrillion dollar IPO? People preferring cheapo models over its flagship AI products. Recent spending data from 70,000 US companies collected by the payments group Ramp shows that spending on Fable 5, Anthropic's priciest and most powerful model, has plateaued at only 11 percent of the overall outlay, or money spent, on its AI tools, the Financial Times reported. It reflects a shift in how enterprise users are using flagship models, reserving them only for the most complex tasks while letting more than serviceable cheaper models take care of the dirty work. Those cheaper alternatives can be Anthropic's own older models, or open-weight models offered by Chinese competitors. If the pattern holds, according to the FT, it could upend the go-big-or-go-home business model of leading AI labs, which have focused on pouring their resources into building even larger and more complex models. "Most people don't need to operate at the frontier," Miles Clements, a partner at the venture capital firm Accel, which has invested $1 billion in Anthropic, told the FT. The period when customers favored using only frontier models "was not a durable era," Clements added. Fable 5 had a rocky launch in June. Its hype was clouded by its purportedly powerful ability to launch cyberattacks autonomously -- a narrative that Anthropic helped fuel, it's worth noting. The Trump administration ordered Anthropic to suspend access to its model to foreign customers, citing the national security risks posed by the model, but later lifted the export restrictions. The hope that take-up of Fable 5 would accelerate once the political controversy cleared hasn't been borne out, however, with the Anthropic model's adoption lagging behind the releases of its previous frontier models. Its annualized revenue -- the amount it's projected to make in a year based on its current performance -- in July reached $65 billion, which is well short of the $80 billion estimate set by bullish investors, the FT noted. OpenAI, meanwhile, is nipping at Anthropic's heels. After losing significant ground to Anthropic this year, the ChatGPT maker's annualized revenue has surged to $40 billion, with its new GPT-5.6 model, which is much cheaper to use than Fable 5, boosting sales. Some experts are seeing the plateaued Fable 5 spending as a glass half-full. Alex Imas, director of AGI economics at Google DeepMind, argued that Anthropic isn't worried about Fable spending in isolation but the "total spend across all models."
Cerebras CS-4 Delivers A 30x Uplift In AI This Year, But Next-Gen Racks Already In Works With CS-5 Pumping Out 10K TPS Per User In 2027 & CS-6 Bringing 3D Wafer-Scale SRAM
Cerebras is rolling out its CS-4 AI rack-scale solution this year but is also working on its next-gen CS-5 and CS6 solutions. Cerebras Takes Wafer Scale Engine To The Next-Level With CS-4, CS-5, and CS-6 AI Racks Last week, Cerebras took the curtains off its CS-4 rack-scale solution, which is powered by the WSE-3T chip. The WSE-3T is a boosted version of the WSE-3 (Wafer Scale Engine), offering much higher capabilities. At Hot Chips 2026, Cerebras is providing a deeper dive into its rack-scale solutions while also giving us a look at its next-gen solutions. So starting off with the Cerebras CS-4, it offers up to a 2x uplift in token generation speed and up to a 10x uplift in throughput per watt versus Cerebras's CS-3 solution. Being much larger than any GPU or compute accelerator due to its wafer-scale design, the current-gen WS-3 is already much ahead of the aforementioned solutions in Token Generation speed. With CS-4, that lead is taken up to 30x across various models. One of the advantages of the wafer-scale design is that Cerebras has a lot of room to expand. The chip is massive in size versus today's leading AI accelerator, NVIDIA Rubin. A single NVIDIA Rubin chip delivers up to 22 TB/s memory bandwidth, and AMD's MI455X delivers up to 23.3 TB/s of raw memory bandwidth. Cerebras CS-4 with a single WSE-3T chip offers 43,200 TB/s bandwidth, which is 2000x more bandwidth than Rubin. But we also need to understand that the bandwidth figures for NVIDIA and AMD chips are based on HBM4 solutions, while Cerebras is measuring the raw BW offered by the SRAM onboard the wafer. Moving forward, Cerebras is integrating the WSE-3T chips on CS-4 within its nexus rack-scale platform. Compared to CS-3, the new solution is designed with modularity in mind, offering a simpler build, faster deployment, and independent integration of power, compute, and IO. Nexus Goes Modular With WSE Backpacks & A Wire-Free Package Each Nexus rack is attached to pluggable backpacks, which contain one WSE-3T chip, each. Each of these backpacks is an innovation on its own, featuring a wafer package attached to twice the cooling and power, a wafer IO module with twice the bandwidth, 2x faster latency and room for additional modular upgrades, and the whole pack leverages an efficient manufacturing process which uses 50% fewer components and is 60% automated versus CS-3. Power distribution is important for AI factories. Each loss can lead to a severe lack of compute output. On Rubin, it is claimed that the 50mm distance between the converters and silicon leads to major power losses along paths within the PCB. As such, the system requires more copper layers to reduce resistive losses, which add to the costs and complexity of the system. Cerebras bypasses these losses through its 54.5VDC Busbar which involves no PCB between the power and the chip. The chip sits directly on the DC/DC Power convertors, and then there's also the shorter distance between the array of AC/DC power converters, which reduces extra resistive power losses and parasitic inductances. The result is a 100x improvement versus a traditional GPU setup since the distances are cut down to just 0.5mm (vs 50mm). For IO, Cerebras makes use of a next-gen interface module which extends the fabric from wafer edges and is both modular and programmable for the future. The IO interface is also low latency and high bandwidth, made possible through new direct wafer link interfaces and a standard RoCE protocol network. Each backpack also includes integrated water conditioning. The pack houses a flow regulation actuator which guarantees proper wafer flow rate, a leak detection module, valved dry quick disconnects, and an energy meter, making it easy to install while monitoring the system for potential leaks. The cooling itself goes in the rear. On the front, Cerebras houses the power, which includes AC/DC PSUs with up to 277VAC input and 54.5VDC output. There are up to 30x PSU modules per backpack, so 90 in total, and all of these are air-cooled with dedicated fan modules for cooling smaller devices in the backpack. Another area in which Cerebras shows its rack prowess is the fabric. CS-4 features a 53.5 PB/s fabric on the wafer itself and has no need for cables. Meanwhile, NVIDIA's Rubin NVL72 racks feature 5000 cables, offering up to 260 TB/s of NVLINK fabric speeds. This means CS-4 offers 200x higher fabric bandwidth than GPU interconnects. The bandwidth and latency advantages don't stop at the fabric. The WSE-3T chip offers 2.4 Tb/s of aggregate bandwidth at 3us latency, while the network latency between the wafers is 1.7x higher. We then move to a generalized comparison between the CS-3 and CS-4 rack-scale solutions. CS-3 was capable of 125 PFLOPs on a single WSE-3 chip while CS-4 offers 750 PFLOPs of AI compute with three WSE-3T chips (250 PFLOPs per chip). It has 132 GB of SRAM (44 GB per chip) versus 44 GB on the previous rack, with much higher bandwidth and less than half the bandwidth. With the speeds and feeds done, Cerebras showcases what to expect in terms of AI compute while also highlighting the capabilities of its current CS-3 rack, which it claims already runs the largest frontier model (GPT-5.6 SOL @ 10T parameters). Lastly, for the CS-4, Cerebras has already announced that the rack is in early access and general availability is scheduled for Q3 2026. CS-5 Tackles The AI Wall With Boosted Capabilities While CS-6 Goes 3D But there's more: Cerebras is also announcing its next-gen CS-5 and CS-6 rack-scale solutions for the first time. According to Cerebras, CS-5 will be launching in 2027 and is "Designed to set another standard of speed and efficiency". This solution will scale from 30B to multi-trillion-parameter models. For Gemma 4 31B and gpt-oss 120B, the company estimates up to 10,000 tokens per second per user, and in frontier models such as DeepSeek, Kimi, GPT 5.6 SOL, CS-5 is expected to hit up to 5000 tokens per second per user with up to 3 million tokens per second per MW. There's also CS-6, which is expected to take full advantage of 3D packaging solutions with a yield-resilient architecture, a vertical power delivery solution, & a fully integrated cooling methodology. The Wafer Scale Engine for CS-6 will integrate Wafer-Scale SCRAM on top of the WSE chip through 3D integration, while being an order of magnitude smaller in footprint with the fastest AI inference speeds on the market. This is a very forward-looking design for now, but it looks like Cerebras has the stage set for future AI models and is scaling its wafer-scale engines to meet the accelerated AI growth big time. Follow Wccftech on Google to get more of our news coverage in your feeds.
Dow Jones Top Company Headlines at 5 PM ET: OpenAI's Head of Data Centers Has Left the Company | Anthropic ...
OpenAI's Head of Data Centers Has Left the Company Chris Malone joins a string of recent high-level executive departures as the AI giant heads toward an IPO and ramps up its spending on computing power. ---- Anthropic Expected to Tell Investors It Sees Over $30 Trillion in Potential Revenue The AI startup is likely to top SpaceX's eye-popping potential revenue estimate. ---- California Attorney General Ramps Up Criticism of Paramount Rob Bonta said Paramount is more focused on the "court of public opinion" than the court of law in the Warner merger fight. ---- Intuit Forecasts Slower Growth, Takes Steps to Win More TurboTax Users The company said it expects revenue to increase 9% to 10% for fiscal 2027, slowing from 14% this year, as it recorded a lower profit in its latest quarter. ---- Zoom Reports Solid Earnings. The Stock Drops Anyway. Zoom reported adjusted earnings of $1.55 a share for the quarter, up from $1.53 a year ago and above analysts' expectations for $1.48, per FactSet. ---- Starbucks Union Calls for Boycott in Latest Test of Company Turnaround Starbucks Workers United is seeking a contract agreement. For investors, the key question is whether the boycott move will affect customer behavior at a meaningful scale. ---- Dick's Sporting Goods Bet Big on Sneakers With Foot Locker. It Backfired. Shares tumbled after the retailer revealed footwear discounts are sapping profits. The chairman defended a $2.4 billion deal for the sneaker chain. ---- United Airlines Ups Its Bet on Instagrammable International Destinations Robust travel demand from Americans is helping fuel new routes like Marseille and Ibiza. ---- Bank of Montreal Plans to Buy Back Shares After Strong Quarter The big Canadian lender benefited from double-digit revenue growth and an improved credit performance in its fiscal third quarter, with the lowest provision on impaired loans in the last 10 quarters, though its bottom line was squeezed by one-time charges for its exit from certain businesses. ---- Scotiabank Earnings Lifted by Record Result in Wealth Management Bank of Nova Scotia notched a rise in third-quarter earnings, driven by strong results across its business lines that included a record result from its global wealth management and global banking and markets operations. ---- OnlyFans Paid $700 Million Dividend to Founder Year Before He Died The online platform, widely known for its explicit content, was a reliable cash machine despite employing fewer than 50 people. ---- Oura and Dunkin' Get Ready to Join IPO Bonanza Investors are sizing up the parade of new offerings expected in coming months. ---- Hyundai Motor, Union Reach Wage Deal to End Walkouts The tentative wage agreement ends months of on-and-off walkouts that have disrupted production at the Korean automaker. ---- One of Taiwan's Biggest AI Winners Isn't a Chip Maker Taiwan's richest person rode the AI boom but his company has nothing to do with technology-it's a furniture-component maker called King Slide Works. ---- Woodside Puts 'Everything on the Table' in Review of Beaumont Facility Chief Executive Liz Westcott said the review reflected a shift in international policy positions and slack demand for lower carbon ammonia. ---- Lego Sales Growth Outpaces Global Toy Market to Hit New Record Lego reported record sales for the first half of the year as it launched hundreds of new products. (END) Dow Jones Newswires August 25, 2026 17:15 ET (21:15 GMT) Copyright (c) 2026 Dow Jones & Company, Inc.
Kraken Pro Goes Live on TradingView for Direct Terminal Trading - Crypto Economy
Kraken Pro has integrated as an official broker on TradingView, allowing users to execute buy and sell orders directly from the analysis terminal without needing to switch windows or platforms. The connection is established via OAuth, a process that takes less than one minute and does not require the generation, storage, or rotation of API keys. Once the account is linked, TradingView's broker panel displays the available balance, position size, and order types without leaving the chart. The integration covers the spot market across the full range of pairs available on Kraken Pro. From the panel, users can place market, limit, and stop orders, with direct execution against Kraken's order book. Additional products, such as instruments outside the spot segment, will be incorporated in later phases and currently require operating directly from the Kraken Pro terminal. The initiative aims to eliminate friction between technical analysis and execution, a common workflow issue for active traders who use TradingView as their primary analysis environment. Access can be revoked at any time from the account settings on the exchange. Source: https://blog.kraken.com/product/pro/now-available-on-tradingview Disclaimer: Crypto Economy Flash News are based on verified public and official sources. Their purpose is to provide fast, factual updates about relevant events in the crypto and blockchain ecosystem. This information does not constitute financial advice or investment recommendation. Readers are encouraged to verify all details through official project channels before making any related decisions.
GTA 6 Leak Subpoena: Discord Says It Will 'Evaluate the Validity and Scope Before Responding'
Related reads:[UPDATE] GTA 6 Leaker Releases Game Store, Nightclub and Beach Gameplay Videos, Defends Crypto Scheme Discord has countered online reports about Take-Two's hunt for the identity of the GTA 6 leaker, saying it has yet to be formally served a subpoena, and when it does receive it, it will take a long, hard look at it before responding. On Friday, Andrew L. Carter Jr., Judge of the United States District Court for the Southern District of New York, granted Take-Two's request to subpoena Discord to divulge identifying information about users the Rockstar owner believes are tied to the leaks. Take-Two's subpoena of Microsoft was also granted. But while Microsoft has issued a statement to say it's happy to work with Take-Two and Rockstar to help unearth the GTA 6 leaker, Discord issued a different line overnight. "I can share an update on this," Discord marketing director Ryan K. Rigney said in a tweet. "Discord has not yet been served with a subpoena from Take-Two. When we do, we'll evaluate the validity and scope before responding." Take-Two issued the subpoenas alleging Digital Millennium Copyright Act (DMCA) infringement. It wants information and records relating to the 'CyberLeek' account, which is allegedly behind the GTA 6 leaks. The subpoenas request identifying information associated with user accounts that were members of a number of Discord servers. One of the Discord servers mentioned in the filing is 'discord.gg/darkviperau,' which is described as the DarkViperAU editors' server. DarkViperAU is the online name of Australian GTA content creator Matthew "Matt" Judge, who has 2.27 million subscribers on YouTube and nearly 900,000 followers on Twitch. Judge has denied any wrongdoing. Reclaim The Net, a group that advocates for free speech and privacy, criticised the Discord subpoena in a tweet. "Copyright law is supposed to target infringement," it said. "Here, it's being used to hunt for identities. Take-Two is seeking DMCA subpoenas that could expose users who didn't leak or upload GTA VI footage, but who participated in conversations within Discord servers where the leak was discussed." In an article, Reclaim The Net continued: "One of the Discord servers in question has over 107,000 members, and while it is not clear if Take-Two is going after each and every one of them, the request does say 'all identifying information associated with all user accounts' that were or are members of three named servers since June 1 - and that participated in a conversation in those servers. "In this case, Take-Two Interactive's one filings make it clear that they're not limited to whoever is uploading the leaks, they're targeting entire communities and channels that are merely existing in the same Discord server as those alleged leakers. "This is now the pursuit of a membership list under the color of the enforcement of copyright. While it can be crushed if people push back, there's actually nothing about the DMCA process that discourages this in a strong enough way. "To file a claim is cheap and quick, and the compliance cost falls on the platforms. There was a time when companies would at least try to maintain the guise that their targets were limited to infringement but, these days, they're getting more brazen and now hitting anyone who happens to have been standing nearby." It is unclear if Microsoft has actually been served with Take-Two's subpoena, either, although according to court documents it looks like Microsoft had already launched an investigation of its own into the CyberLeek account. Meanwhile, the subpoena for Twitter / X has yet to be signed off, but that may simply be a formality. All the while, the GTA 6 leaks continue, and there is fresh concern that CyberLeek will release significant story spoilers. Neither Rockstar nor Take-Two have issued a statement. IGN has asked for comment. GTA 6 is due out on PlayStation 5 and Xbox Series X and S on November 19, 2026. Photo by Jakub Porzycki/NurPhoto via Getty Images. Wesley is Director, News at IGN. Find him on Twitter at @wyp100. You can reach Wesley at [email protected] or confidentially at [email protected]. Related reads:Former Rockstar Dev Calls GTA 6 Leaks 'A Nothing Burger,' Insists Hot Coffee Was Much, Much Worse
Anthropic Asks Job Candidates: How Much Money Do You Have in the Bank?
Anthropic has introduced a direct and sometimes uncomfortable financial inquiry during its hiring process for senior roles. According to a report from Axios, recruiters now ask prospective employees a pointed question about their personal financial situation: how much money they have in the bank and whether they could afford to work for a year without drawing a salary. This approach marks a shift in how one of the leading artificial intelligence companies evaluates talent. The query serves multiple purposes. First, it helps determine a candidate's genuine interest in the mission of building safe and reliable AI systems rather than chasing the highest compensation package available in a competitive market. Second, it signals the company's preference for individuals who demonstrate financial independence and long-term commitment over those who might treat the position as a short-term stepping stone. The practice reflects broader pressures facing AI organizations as they scale rapidly. Anthropic, valued at more than $60 billion following recent funding rounds, competes fiercely with OpenAI, Google DeepMind, Meta, and numerous well-funded startups for the same limited pool of researchers, engineers, and policy experts. Compensation packages in this sector often include seven-figure salaries, significant equity grants, and performance bonuses that can reach tens of millions of dollars. Against that backdrop, a question about personal savings can feel jarring. Candidates who have encountered the inquiry describe it as blunt but effective at revealing priorities. Some report being asked variations of the question during late-stage interviews, typically after technical assessments and team meetings have already taken place. Recruiters frame the discussion around alignment with company values, emphasizing that Anthropic seeks people motivated by the potential societal impact of advanced AI rather than purely financial gain. The company's leadership has long stressed the importance of careful, responsible development of frontier AI models. Dario Amodei, Anthropic's chief executive and co-founder, has spoken publicly about the need to prioritize safety research even when it slows commercial progress. This philosophy appears to extend to hiring decisions. By probing financial circumstances, the organization aims to identify individuals who share that patient, mission-driven outlook and who will not be easily lured away by competing offers. Industry observers point out that such questions, while uncommon in most corporate settings, have precedents in certain specialized fields. Venture capital firms sometimes evaluate founders based on their runway and personal commitment. Research institutions have historically favored academics who demonstrate dedication to pure inquiry over monetary rewards. In the AI sector, where talent wars have driven compensation to extraordinary levels, Anthropic's approach represents an attempt to filter for intrinsic motivation. Not every candidate responds positively to the question. Some view it as an invasion of privacy that has little bearing on their professional qualifications. Others worry that answering honestly could weaken their negotiating position on salary and equity. Legal experts note that while employers generally have latitude to ask about financial stability in certain contexts, such questions must be applied consistently to avoid potential discrimination claims. Anthropic appears to direct the inquiry primarily at senior individual contributors and leadership positions rather than entry-level roles. The timing of this reported practice coincides with significant changes in the AI industry funding environment. After years of abundant capital and skyrocketing valuations, investors have grown more selective about where they deploy resources. Companies face pressure to demonstrate efficient growth and sustainable business models. In this climate, organizations like Anthropic may see advantages in building teams of people who are less likely to demand constant compensation increases or depart for marginally better offers elsewhere. Former employees and recruiters familiar with the company's process suggest the financial question forms part of a larger evaluation framework. Interviewers also explore a candidate's views on AI ethics, their tolerance for uncertainty in a rapidly changing field, and their willingness to engage with complex safety challenges. The combination of technical excellence, philosophical alignment, and financial independence appears to define the ideal Anthropic profile. This hiring strategy carries both advantages and risks. On the positive side, it may help create a more stable workforce less susceptible to the frequent job-hopping that characterizes Silicon Valley. Employees who join primarily for the work itself often display higher engagement and remain through challenging periods. The approach could also foster a culture where decisions prioritize long-term safety considerations over short-term commercial gains. Potential drawbacks exist as well. The question could inadvertently screen out talented individuals who carry student debt, support families, or come from backgrounds without generational wealth. In an industry already criticized for lacking diversity, additional financial filters might narrow the applicant pool further. Some critics argue that true commitment should be assessed through past behavior, research contributions, and interview performance rather than personal balance sheets. Anthropic has not publicly commented on the specific hiring question, but its overall approach to recruitment emphasizes careful selection. The company maintains a relatively small headcount compared to its valuation and ambitions, suggesting a deliberate focus on quality over quantity. This selectivity extends beyond technical skills to encompass character traits and personal circumstances that might influence an employee's staying power. The broader AI talent market continues to evolve in response to these dynamics. Other organizations have adopted different strategies to attract and retain top performers. Some offer equity packages tied to multi-year vesting schedules with steep cliffs. Others emphasize prestigious research environments and the opportunity to publish groundbreaking work. A few have experimented with profit-sharing models or impact-focused incentives designed to appeal to mission-driven candidates. For job seekers in artificial intelligence, the emergence of such questions requires new preparation. Candidates must consider not only how to present their technical accomplishments but also how to articulate their personal motivations and financial resilience. Those uncomfortable discussing their savings may need to weigh whether a particular organization's culture aligns with their own values and boundaries. The practice also raises interesting questions about the relationship between personal wealth and professional dedication. Does financial independence truly correlate with better performance in high-stakes AI development? Or does it simply reflect a preference for candidates from privileged backgrounds? These debates touch on larger societal conversations about merit, opportunity, and the role of money in shaping technological progress. As artificial intelligence capabilities advance, the humans guiding that development take on increasing significance. Companies like Anthropic appear to believe that understanding a candidate's complete picture, including their financial situation, provides valuable insight into their potential contributions. Whether this approach proves successful will become clearer over time as the organization releases new models and navigates the complex challenges of scaling safe AI systems. The trend toward more personal and values-based hiring criteria may spread beyond Anthropic. In an industry where traditional metrics like degrees from elite universities or publications in top conferences no longer sufficiently distinguish candidates, organizations are searching for additional signals of fit. Financial questions represent one such signal, albeit a controversial one that forces both companies and candidates to confront the role of money in ostensibly mission-driven work. Recruiting professionals predict that similar inquiries could appear at other AI laboratories facing comparable pressures. The combination of high burn rates, intense competition, and existential questions about technology governance creates conditions where conventional hiring practices feel inadequate. Organizations may increasingly look beyond resumes to assess the whole person, including their economic circumstances and underlying motivations. For now, the reported Anthropic practice stands out as a notable example of how far some companies will go to ensure alignment between their ambitious goals and the individuals tasked with achieving them. The question about bank balances serves as both a practical assessment of runway and a philosophical litmus test. In an era of unprecedented investment in artificial intelligence, it reminds everyone involved that the most valuable resource remains committed human attention guided by something deeper than financial reward. This development occurs against a backdrop of growing scrutiny over AI company practices, from compensation structures to safety commitments. How organizations answer the question of what makes a good AI researcher or engineer will influence not only their competitive position but also the direction of technological development itself. Anthropic's willingness to ask uncomfortable financial questions suggests a conviction that getting the right people matters more than maintaining conventional recruiting etiquette. As the field matures, other companies may find themselves adopting or adapting similar approaches to secure the talent they believe will determine success in the coming years of AI advancement.
Anthropic Asks Job Candidates: How Much Money Do You Have in the Bank?
Anthropic has introduced a direct and sometimes uncomfortable financial inquiry during its hiring process for senior roles. According to a report from Axios, recruiters now ask prospective employees a pointed question about their personal financial situation: how much money they have in the bank and whether they could afford to work for a year without drawing a salary. This approach marks a shift in how one of the leading artificial intelligence companies evaluates talent. The query serves multiple purposes. First, it helps determine a candidate's genuine interest in the mission of building safe and reliable AI systems rather than chasing the highest compensation package available in a competitive market. Second, it signals the company's preference for individuals who demonstrate financial independence and long-term commitment over those who might treat the position as a short-term stepping stone. The practice reflects broader pressures facing AI organizations as they scale rapidly. Anthropic, valued at more than $60 billion following recent funding rounds, competes fiercely with OpenAI, Google DeepMind, Meta, and numerous well-funded startups for the same limited pool of researchers, engineers, and policy experts. Compensation packages in this sector often include seven-figure salaries, significant equity grants, and performance bonuses that can reach tens of millions of dollars. Against that backdrop, a question about personal savings can feel jarring. Candidates who have encountered the inquiry describe it as blunt but effective at revealing priorities. Some report being asked variations of the question during late-stage interviews, typically after technical assessments and team meetings have already taken place. Recruiters frame the discussion around alignment with company values, emphasizing that Anthropic seeks people motivated by the potential societal impact of advanced AI rather than purely financial gain. The company's leadership has long stressed the importance of careful, responsible development of frontier AI models. Dario Amodei, Anthropic's chief executive and co-founder, has spoken publicly about the need to prioritize safety research even when it slows commercial progress. This philosophy appears to extend to hiring decisions. By probing financial circumstances, the organization aims to identify individuals who share that patient, mission-driven outlook and who will not be easily lured away by competing offers. Industry observers point out that such questions, while uncommon in most corporate settings, have precedents in certain specialized fields. Venture capital firms sometimes evaluate founders based on their runway and personal commitment. Research institutions have historically favored academics who demonstrate dedication to pure inquiry over monetary rewards. In the AI sector, where talent wars have driven compensation to extraordinary levels, Anthropic's approach represents an attempt to filter for intrinsic motivation. Not every candidate responds positively to the question. Some view it as an invasion of privacy that has little bearing on their professional qualifications. Others worry that answering honestly could weaken their negotiating position on salary and equity. Legal experts note that while employers generally have latitude to ask about financial stability in certain contexts, such questions must be applied consistently to avoid potential discrimination claims. Anthropic appears to direct the inquiry primarily at senior individual contributors and leadership positions rather than entry-level roles. The timing of this reported practice coincides with significant changes in the AI industry funding environment. After years of abundant capital and skyrocketing valuations, investors have grown more selective about where they deploy resources. Companies face pressure to demonstrate efficient growth and sustainable business models. In this climate, organizations like Anthropic may see advantages in building teams of people who are less likely to demand constant compensation increases or depart for marginally better offers elsewhere. Former employees and recruiters familiar with the company's process suggest the financial question forms part of a larger evaluation framework. Interviewers also explore a candidate's views on AI ethics, their tolerance for uncertainty in a rapidly changing field, and their willingness to engage with complex safety challenges. The combination of technical excellence, philosophical alignment, and financial independence appears to define the ideal Anthropic profile. This hiring strategy carries both advantages and risks. On the positive side, it may help create a more stable workforce less susceptible to the frequent job-hopping that characterizes Silicon Valley. Employees who join primarily for the work itself often display higher engagement and remain through challenging periods. The approach could also foster a culture where decisions prioritize long-term safety considerations over short-term commercial gains. Potential drawbacks exist as well. The question could inadvertently screen out talented individuals who carry student debt, support families, or come from backgrounds without generational wealth. In an industry already criticized for lacking diversity, additional financial filters might narrow the applicant pool further. Some critics argue that true commitment should be assessed through past behavior, research contributions, and interview performance rather than personal balance sheets. Anthropic has not publicly commented on the specific hiring question, but its overall approach to recruitment emphasizes careful selection. The company maintains a relatively small headcount compared to its valuation and ambitions, suggesting a deliberate focus on quality over quantity. This selectivity extends beyond technical skills to encompass character traits and personal circumstances that might influence an employee's staying power. The broader AI talent market continues to evolve in response to these dynamics. Other organizations have adopted different strategies to attract and retain top performers. Some offer equity packages tied to multi-year vesting schedules with steep cliffs. Others emphasize prestigious research environments and the opportunity to publish groundbreaking work. A few have experimented with profit-sharing models or impact-focused incentives designed to appeal to mission-driven candidates. For job seekers in artificial intelligence, the emergence of such questions requires new preparation. Candidates must consider not only how to present their technical accomplishments but also how to articulate their personal motivations and financial resilience. Those uncomfortable discussing their savings may need to weigh whether a particular organization's culture aligns with their own values and boundaries. The practice also raises interesting questions about the relationship between personal wealth and professional dedication. Does financial independence truly correlate with better performance in high-stakes AI development? Or does it simply reflect a preference for candidates from privileged backgrounds? These debates touch on larger societal conversations about merit, opportunity, and the role of money in shaping technological progress. As artificial intelligence capabilities advance, the humans guiding that development take on increasing significance. Companies like Anthropic appear to believe that understanding a candidate's complete picture, including their financial situation, provides valuable insight into their potential contributions. Whether this approach proves successful will become clearer over time as the organization releases new models and navigates the complex challenges of scaling safe AI systems. The trend toward more personal and values-based hiring criteria may spread beyond Anthropic. In an industry where traditional metrics like degrees from elite universities or publications in top conferences no longer sufficiently distinguish candidates, organizations are searching for additional signals of fit. Financial questions represent one such signal, albeit a controversial one that forces both companies and candidates to confront the role of money in ostensibly mission-driven work. Recruiting professionals predict that similar inquiries could appear at other AI laboratories facing comparable pressures. The combination of high burn rates, intense competition, and existential questions about technology governance creates conditions where conventional hiring practices feel inadequate. Organizations may increasingly look beyond resumes to assess the whole person, including their economic circumstances and underlying motivations. For now, the reported Anthropic practice stands out as a notable example of how far some companies will go to ensure alignment between their ambitious goals and the individuals tasked with achieving them. The question about bank balances serves as both a practical assessment of runway and a philosophical litmus test. In an era of unprecedented investment in artificial intelligence, it reminds everyone involved that the most valuable resource remains committed human attention guided by something deeper than financial reward. This development occurs against a backdrop of growing scrutiny over AI company practices, from compensation structures to safety commitments. How organizations answer the question of what makes a good AI researcher or engineer will influence not only their competitive position but also the direction of technological development itself. Anthropic's willingness to ask uncomfortable financial questions suggests a conviction that getting the right people matters more than maintaining conventional recruiting etiquette. As the field matures, other companies may find themselves adopting or adapting similar approaches to secure the talent they believe will determine success in the coming years of AI advancement.
As Cerebras Launches a New, Record-Setting AI Accelerator, Here's How You Should Play CBRS Stock
Cerebras Systems (CBRS) is giving investors another reason to pay attention to its ambitions in the rapidly expanding artificial intelligence (AI) accelerator market. On Aug. 18, the company unveiled its new CS-4 rack-scale platform, which it says can deliver up to 30 times faster AI inference than comparable GPU-based systems. Built around three new WSE-3 Turbo processors, CS-4 delivers 750 petaflops of AI compute, 7.2 terabits per second of I/O bandwidth, and 129.6 petabytes per second of memory bandwidth. The launch comes at a critical time for Cerebras. The company is seeking to establish itself as a credible alternative to Nvidia (NVDA) in AI inference, where demand is rising as businesses deploy increasingly sophisticated generative AI and agentic applications. Cerebras says CS-4 can support models exceeding 50 trillion parameters and reduce wafer-to-wafer latency to as little as two microseconds, potentially giving customers a significant speed advantage for latency-sensitive workloads. More News from Barchart However, the technology opportunity must be weighed against Cerebras' execution challenges. In its second quarter, reported revenue reached $180.1 million. Yet profitability remains a concern, and CBRS shares have shown considerable volatility following the company's recent earnings report. For investors, CS-4 could strengthen the long-term bullish case, but the stock remains a high-risk AI play. About Cerebras Systems Stock Cerebras Systems is a Sunnyvale, California-based artificial intelligence semiconductor company that develops specialized computing systems and processors designed to accelerate AI workloads, particularly inference. Its flagship Wafer-Scale Engine (WSE) technology integrates compute and memory on a single wafer, offering an alternative to conventional GPU-based architectures. The company has a market cap of around $49.1 billion. Cerebras has experienced significant volatility since its Nasdaq debut, as investor excitement over the AI infrastructure opportunity has been tempered by concerns surrounding its valuation and profitability. The company priced its IPO at $185 per share and started trading on May 14, 2026. CBRS opened at $350 and ended its first trading session at $311.07, marking a 68.2% gain over its IPO price and placing it among the year's strongest new listings.
AI giant Anthropic in talks to lease massive Bastrop County data center, sources said
by: Justin Sayers- Senior Staff Writer, Austin Business Journal AUSTIN (ABJ) -- Fresh off its huge data center lease in Milam County northeast of Austin, Silicon Valley-based Anthropic PBC is in talks to occupy a gargantuan data center development on the east side in Bastrop County. While no agreement has been finalized, the artificial intelligence company has been working with Washington, D.C.-based BlackChamber Partners LLC and Dallas-based Pacifico Energy LLC on what's been described as a 2,842-acre gas-fired electric generation facility to support a major data center development in the Cedar Creek area, several sources told ABJ. Read the full article on the Austin Business Journal.
Companies not as keen on Anthropic's best AI model
Enterprises appear to be prioritizing cheaper AI tools instead. endif; ?> New payment data from Ramp suggests that Anthropic's most advanced AI model, Fable 5, has had a slow start among enterprise customers, the Financial Times reports. Two months after launch, Fable 5 accounts for about 11% of total enterprise spending on Anthropic models, breaking the previous trend of customers quickly moving to the most powerful model. According to analysts and investors, the development is mainly due to Fable 5's high price and the fact that cheaper models are good enough for most tasks.
Anthropic upgrades Claude's streaming renderer, claims 9x fewer stalls on slower laptops
The frontend optimization delivers smoother token streaming for long responses without changing Claude's underlying models or API Anthropic rolled out a frontend optimization to Claude's streaming response renderer on August 24, targeting a problem that anyone who's tried to get a long answer from an AI chatbot on an older machine knows well: the maddening stutter-and-freeze cycle that turns a conversation into a slideshow. The update applies to Claude's web and desktop applications. According to Anthropic, long answers now stream roughly 4x smoother, stalls on slower laptops drop by 9x, and worst-case freezes shrink by 4.5x. On hardware that supports it, like 120 Hz MacBooks, the renderer can sustain 120 fps during streaming output. What actually changed under the hood The core fix is elegant in its simplicity. Previously, each incoming token from Claude's model triggered a repaint of the full response container, or at least a substantial portion of it. The new approach restricts UI updates to only the elements that have actually changed, resulting in dramatically less computational overhead per token, which matters most on machines with limited GPU headroom or older processors. Critically, Anthropic made no changes to its models or API with this release. Generation speed, output quality, and the underlying inference pipeline remain untouched. This is purely a presentation-layer improvement. Why UI performance matters in the AI race The 120 fps sustained frame rate figure is particularly telling. Most web applications don't need to think about frame rates at all, but streaming text renderers are effectively animations. Each new token is a frame update. When you're generating hundreds of tokens for a long response, that's hundreds of sequential frame updates, and any hitch becomes visible as a stutter or freeze. Hitting 120 fps on compatible hardware means Anthropic's renderer can now match the refresh rate of Apple's ProMotion displays without dropping frames. The feedback loop, and its limits User reception on social media was largely positive, with many noting that the improvement was immediately noticeable during extended conversations. But some responses highlighted a tension that Anthropic and every other AI company faces: polishing the interface doesn't fix the model. Several users pointed out that smoother streaming doesn't help when Claude hallucinates a citation or confidently delivers incorrect information. The update also has implications for Anthropic's enterprise ambitions. Corporate users often run standardized hardware that skews older than what developers and early adopters use. A 9x reduction in stalls on slower laptops isn't just a nice stat for a blog post. It's the difference between an enterprise deployment that employees actually use and one they abandon for a competitor after a week of frustrating freezes.
Nvidia reportedly eyes another investment in Perplexity AI at a $30B valuation
Nvidia reportedly eyes another investment in Perplexity AI at a $30B valuation Nvidia Corp. is reportedly considering making another investment in the artificial intelligence search startup Perplexity AI Inc. A report by The Information says the chipmaker is holding talks with Perplexity over an investment that could push the startup's valuation to more than $30 billion. That would represent a jump of more than 50% from the $20 billion valuation Perplexity finalized about a year ago, when it last raised money. The size of Nvidia's potential investment was not disclosed, and there's no guarantee that any deal would be reached, The Information said, citing anonymous sources who are familiar with the discussions. Neither Nvidia nor Perplexity would comment on the reported discussions. Perplexity is an attractive target for investors for its business has continued to grow at a rapid rate. According to The Information, the startup has grown its annualized revenue run rate to an impressive $750 million, up from less than $250 million at the start of the year. If true, that would mean it has managed to triple its revenue run rate in just eight months. One of the main reasons for that impressive growth is Perplexity Computer, a cloud-based AI agent that was first released in April for Mac computers and later expanded to Windows devices. Perplexity Computer is designed to automate computer tasks for professional users. It acts as a general-purpose digital worker that can access authorized files and applications on a user's computer. Users can ask it to create or edit Word documents, update Excel spreadsheets, organize files, conduct online research and complete workflows involving multiple applications. The proposed investment would deepen an existing relationship between Nvidia and Perplexity. The chipmaker is already one of its main financial backers, alongside Amazon.com Inc. founder Jeff Bezos and SoftBank Group Corp. Nvidia has become an increasingly important partner for AI startups like Perplexity, and sees its bet on the startup as an investment in its future. As the world's top supplier of silicon for high-frequency AI inference, it has a vested interest in making sure that the search layer - which is a massive compute ecosystem - remains aligned with its chip ecosystem. What Nvidia doesn't want is for the likes of Perplexity and others to go sniffing around rival chipmakers such as Advanced Micro Devices Inc. and Cerebras Systems Inc., which both offer alternative chips for AI inference. In that way, Nvidia is investing in Perplexity as a kind of insurance policy to safeguard its future revenue stream against possible shifts in AI search architecture. Perplexity's strategic importance to Nvidia is amplified by its distribution efforts, such as its integration with Samsung Electronics Co. Ltd.'s Bixby assistant, which brings its search capabilities to around 800 million devices globally. The AI search firm is also believed to be fixed on a 2028 initial public offering, which means Nvidia has a clear timeline to realize a return on its investment. Nvidia's broader portfolio includes many of its major compute customers, including OpenAI Group PBC, Anthropic PBC, SpaceX Corp.'s xAI, Poolside Inc. and Safe Superintelligence Inc. It shows how the chipmaker has taken a systematic approach to ensuring its market dominance. By supplying the critical infrastructure and acting as a key investor at the application layer, Nvidia has effectively built a self-reinforcing cycle of demand for its chips. Nvidia is also trying to provide direct funding to customers that need to invest in its AI compute hardware. It recently struck a deal with six of Wall Street's biggest financial institutions to provide more than $500 billion in financing for AI infrastructure projects, including its own and those of its customers. Meanwhile, Perplexity has been racing to build out the infrastructure foundation it needs to support its own growth. Earlier this year, it struck a $750 million deal with Microsoft Corp. that will see it adopt that company's Azure cloud services to help run its AI workloads.
Anthropic Claude AI Predicts What Ethereum Needs to Hit $4,000
Disclaimer: Crypto is a high-risk asset class. This article is provided for informational purposes and does not constitute investment advice. You could lose all of your capital. Can a single upgrade close a gap of more than 50%? That is the question behind the latest Claude AI price prediction, where the model predicts Ethereum (ETH) reaching $3,500 to $4,000 by year's end 2026, with $3,800 as the realistic base case. The chart already moved first. Ethereum price spent July and August pinned near $1,900 before ripping vertically to $2,448 in a matter of days. That pump reset the entire conversation. A market that looked forgotten is suddenly trading 25% above where it sat two weeks ago. The catalyst behind the forecast is Glamsterdam. It is the largest Ethereum upgrade since the Merge, and it went live on the Platåberget public testnet on August 20. Mainnet is scheduled for Q4. Standard Chartered ties its own $4,000 target directly to that timeline. Treasury demand is building alongside it. BitMine added 9,926 ETH on August 17, pushing its holdings to 5.82 million ETH, roughly 4.8% of supply and around $11 billion. The regulatory piece is still open. Fidelity's staking-enabled FETH filing remains pending SEC review. Flows have held up better than sentiment suggested. ETH ETF inflows over 30 days reached $524.3 million even as daily flows flattened. The bear case is about positioning. Long exposure is crowded at 69.6% of Binance accounts, and a Glamsterdam delay could break $1,860 support. That break risks a slide to $1,500. Make Your Prediction Count With $25 For Free on Kalshi Ethereum Price Prediction: Dario Amodei Claude AI Predicts Glamsterdam Reopens the Ceiling The damage here has been severe and slow. Ethereum peaked near $4,860 in September 2025, then spent five months grinding down through every support it built. February 2026 broke everything at once, dumping price to $1,740. March through May offered a weak recovery toward $2,450 that failed, and June sent Ethereum back to $1,500. July and August built a quiet floor near $1,900. That base is exactly what made this week's candle possible. Ethereum closed at $2,448.0, up $25.2 for a gain of 1.04%, with a session range from $2,356.3 to $2,483.6. The modest change tells you the vertical leg is already done, and ETH price is holding its gains. Resistance sits at $2,483.6, then the May swing near $2,450, which is now cleared, then $2,800. Support runs through $2,356 and $2,100, with the $1,860 line being the one that actually matters. RSI reads 78.70 against a signal line at 62.88. The 16-point gap is elevated without being extreme. That is a healthier picture than a runaway spike. Momentum is strong, and the rising signal line beneath suggests the move has structural support rather than pure reflex. Q4 is where this gets settled. Deliver Glamsterdam on schedule, and $3,800 stops looking distant. Supercharge Your Trading in 2026 With BloFin AI Trading Bots Ethereum Is Betting One Upgrade Can Reopen the Ceiling. LiquidChain Is Betting the Bigger Constraint Is Between Chains. Glamsterdam is designed to make Ethereum itself faster, cheaper, and more capable. LiquidChain is targeting a different bottleneck: the fact that even upgraded networks still operate as isolated liquidity islands. Bitcoin, Ethereum, and Solana each have deep pools of capital, but moving between them still means bridges, duplicated deployments, added fees, and fragmented user flows. LiquidChain is building a single execution layer that connects all 3, enabling a single deployment to reach multiple ecosystems without rebuilding the same application chain by chain. That gives LiquidChain a thesis that does not depend on one network winning. It benefits from activity existing across several major ecosystems at once. The presale is currently priced at $0.01493 with just over $948,000 raised. If the next DeFi cycle is driven by capital moving more freely between chains, LiquidChain is still being valued at the stage where relatively modest inflows can matter.
Discord appeals Brazil's livestream suspension, says ANPD lacks authority | MLex | Specialist news and analysis on legal risk and regulation
By Henrique Santiago ( August 24, 2026, 22:01 GMT | Insight) -- Discord has appealed a National Data Protection Agency decision to ban its livestreaming features in Brazil, arguing that only the judiciary has the authority to impose such restrictions under Brazil's child online safety rules. The company also challenged the agency's decision-making process and said the measure affects millions of users while doing little to curb criminal activity online.Discord has asked Brazil's privacy regulator, the National Data Protection Agency (ANPD), to allow it to resume its live broadcasts.... Prepare for tomorrow's regulatory change, today MLex identifies risk to business wherever it emerges, with specialist reporters across the globe providing exclusive news and deep-dive analysis on the proposals, probes, enforcement actions and rulings that matter to your organization and clients, now and in the longer term. Know what others in the room don't, with features including: * Daily newsletters for Antitrust, M&A, Trade, Data Privacy & Security, Technology, AI and more * Custom alerts on specific filters including geographies, industries, topics and companies to suit your practice needs * Predictive analysis from expert journalists across North America, the UK and Europe, Latin America and Asia-Pacific * Curated case files bringing together news, analysis and source documents in a single timeline Experience MLex today with a 14-day free trial.
xAI asks US appeals court to reverse dismissal of trade secret case against OpenAI | MLex | Specialist news and analysis on legal risk and regulation
( August 24, 2026, 22:26 GMT | Official Statement) -- MLex Summary: xAI argues a US judge erred by dismissing the company's lawsuit accusing OpenAI of running a coordinated campaign to poach xAI engineers and acquire trade secrets through them in an opening brief to the US Court of Appeals for the Ninth Circuit. The lower court misapplied the Defend Trade Secrets Act and improperly discounted corroborating allegations from former xAI employees, xAI said. "The district court's decision sets an impossibly high bar for pleading trade secret claims ... supplying a roadmap for bad actors to evade liability," xAI said.See attached brief.... Prepare for tomorrow's regulatory change, today MLex identifies risk to business wherever it emerges, with specialist reporters across the globe providing exclusive news and deep-dive analysis on the proposals, probes, enforcement actions and rulings that matter to your organization and clients, now and in the longer term. Know what others in the room don't, with features including: * Daily newsletters for Antitrust, M&A, Trade, Data Privacy & Security, Technology, AI and more * Custom alerts on specific filters including geographies, industries, topics and companies to suit your practice needs * Predictive analysis from expert journalists across North America, the UK and Europe, Latin America and Asia-Pacific * Curated case files bringing together news, analysis and source documents in a single timeline Experience MLex today with a 14-day free trial.
Commentary: What retail investors need to know before jumping into Anthropic's IPO
Let's address some important questions. First, what will the growth trajectory be? There isn't much financial data available at this stage - we'll have to wait for the prospectus when it's published. What we do know is that revenues are expected to be in the region of US$100 billion this year - up from about US$10 billion in 2025. But it's worth interrogating those numbers more closely. When you develop an AI agent, you pay Anthropic monthly for "tokens" to access its system and enormous computing power. During the past year, computing costs per task have shot up - semiconductor prices and energy costs have all risen sharply and Anthropic has passed these on. But it means a chunk of that impressive revenue growth is simply programmers paying more to advance existing projects. Though not unusual for a new software company, that growth - even with my caveat - is still impressive. Some believe it will carry on for years. Certainly, the market opportunity for AI seems huge. Anthropic estimates that AI is capable of covering more than 80 per cent of tasks in fields including management, business and finance, computing, architecture, law, arts and the media, among others. The company's strength lies in its Claude coding models and developer tools that can take on this work. Don't hang your head in despair and start worrying about your job yet. The theory is fine; the practice so far is very different. What often seems to happen is that the person who has lost 80 per cent of their routine execution tasks now has to spend that time reviewing and debugging the AI tool's work, to catch all the mistakes it makes and step in where it struggles with messy real-life complexity.
Thomson Reuters Bets $40 Million on Its Own Legal AI to Curb Dependence on Anthropic
Thomson Reuters just took a calculated step away from its heavy reliance on outside AI providers. The information services giant launched Thomson-1, its first proprietary large language model. The move signals a shift in how one of the legal industry's biggest players thinks about building and owning the technology that powers its flagship products. Announced this week, the model draws from an open-source foundation developed by Alibaba. Thomson Reuters adapted it through a process its chief technology officer described as realignment. The result? A system trained on decades of the company's own authoritative legal, tax and news content. Early tests show it holding its own against some of the most advanced general-purpose models on the market. But don't mistake this for a full break from Silicon Valley's AI leaders. CoCounsel, Thomson Reuters' marquee AI assistant for lawyers, still leans primarily on Anthropic's Claude. The new model slots in for specific high-volume tasks where domain knowledge delivers a measurable edge. Joel Hron, the company's CTO, made the strategy plain. Business Insider reported Hron's analogy. "Renting a house, you still have a roof over your head, and somebody's taking care of it, and it's great. But you're not building any equity that compounds into something valuable for you long term." The company spent roughly $40 million on compute, talent and specialized training to create that equity. The numbers tell part of the story. Thomson Reuters used less than 10% of its vast proprietary corpus to train the model. Hundreds of subject-matter experts reviewed outputs. They identified failures. They refined the system to prioritize accuracy over pleasing responses. The approach stands in contrast to the race for ever-larger general models. Thomson-1, also referred to as Thomson in company materials, builds on Snowdon. That variant stems from Alibaba's Qwen model. A joint team with Imperial College London spent months adapting it. They focused on ethical safeguards, de-biasing and safety. "There's nothing that necessarily ties us to Qwen," Hron told reporters. The foundation can evolve. Performance claims come with caveats. In benchmarks released by the company in late July, Thomson competed closely with frontier systems. It matched or exceeded Claude Opus 4.8 in some legal reasoning tasks. It outperformed GPT-5.5, Claude Sonnet 5 and Gemini 3.1 Pro across a mix of evaluations. The tests covered instruction following, long context, coding and professional workflows. Yet the comparisons aren't apples to apples. Thomson benefited from test-time scaling and internal retrieval tools linked to Westlaw and Practical Law. Competing models searched the open web in some evaluations. Internal composites formed part of the mix. Still, the results impressed enough academics. One preferred Thomson's responses even when others answered correctly. Citation quality held up against the leaders. The first real test comes inside CoCounsel Legal. Thomson becomes the default for Tabular Analysis. That feature reviews up to 10,000 documents and fields as many as 100 questions about them. High-volume. Structured. Measurable accuracy. Exactly where a purpose-built model should shine. The broader CoCounsel platform, refreshed on August 20, now incorporates Anthropic's Claude Agent SDK for agentic workflows. It plans, reasons and executes across tools. Hron doesn't hide the continued partnership. Thomson Reuters expanded its deal with Anthropic in May. "Our main objective is to make Thomson the model that powers more and more of CoCounsel's capabilities over time," he said. The in-house system supplements rather than supplants. For now. This hybrid strategy reflects hard realities in professional services AI. General models hallucinate. They sycophantically please users. They lack deep context in regulated fields. Thomson trains explicitly to flag uncertainty. It avoids forcing confident answers when data runs thin. That choice reduces certain failure modes critical for lawyers and tax professionals. Retrieval remains essential. The model pulls from verified sources. Outputs link back to Westlaw, Practical Law or Checkpoint where possible. Thomson Reuters stops short of promising line-by-line traceability for every claim. "We wouldn't claim every output can be traced line by line to an exact statute or ruling," the company stated. Professionals still verify. The system simply makes that verification easier. Training on proprietary data changes the control equation. The company shapes tradeoffs between helpfulness and accuracy. It optimizes for professional caution over casual conversation. And it avoids feeding customer data to third-party labs, maintaining strict contractual prohibitions. Recent coverage highlights the nuance. The New Stack noted the $40 million investment focused on post-training and expert evaluation rather than pre-training from scratch. It also detailed how Thomson integrates with agentic systems while preserving retrieval-augmented generation. The model doesn't replace retrieval. It enhances it. The Next Web added color on the Alibaba connection and academic feedback. It quoted researchers who found Thomson's outputs preferable in quality and citation strength. The article also flagged Alibaba's recent moves toward charging heavy users of Qwen, underscoring why owning adaptations matters. Thomson Reuters published its own early benchmarking in July. CTO Joel Hron and Head of AI Research Jonathan Schwarz wrote that capable models no longer emerge solely from frontier labs. One now comes from their own organization. The piece emphasized combination: authoritative content, expert judgment, professional tools and model development. A smaller open-weight version of the model sits on Hugging Face for researchers. Commercialization beyond Thomson Reuters products remains under consideration. Customers won't buy direct access to Thomson-1 today. It lives inside the company's ecosystem, starting with legal research and document tasks. The timing feels deliberate. CoCounsel has scaled to serve hundreds of thousands of professionals. Demand for trustworthy AI in legal and tax grows as regulators tighten rules. Enterprises want options that reduce vendor lock-in and API costs that can run high at scale. Critics will note the $40 million figure pales against the billions poured into frontier labs. Success here hinges on whether domain-specific tuning plus retrieval consistently beats general models on real workflows. Early signs look promising. But benchmarks tell only part of the story. Real adoption will come from measurable productivity gains and risk reduction for law firms and corporate departments. Thomson Reuters isn't alone in this push. Other data-rich incumbents eye similar paths. The difference lies in execution. Decades of curated content. Teams of domain experts. A product portfolio already embedded in professional routines. Those assets turn training data into a durable advantage. Hron's house-buying metaphor lingers. Renting powerful models delivers immediate capability. Owning one, even if built on someone else's foundation, creates long-term optionality. The company can iterate faster on its own data. It can tune for fiduciary-grade standards. It can expand across tax, compliance and news without renegotiating every capability. CoCounsel's latest agentic features show the complementary play. Built on Claude's SDK, the system orchestrates complex legal tasks. Thomson handles the heavy lifting on document volume and structured analysis. The combination aims for something greater than either alone. Questions remain about geopolitical angles. Reliance on a Chinese open-source base, even heavily adapted, invites scrutiny in some markets. Thomson Reuters stresses the realignment process and independence going forward. Nothing locks them to Qwen. Future versions could draw from other bases or further internal development. For industry watchers, this launch marks a maturation point. Pure reliance on API calls to Anthropic, OpenAI or Google gives way to selective ownership. The $40 million bet tests whether incumbents with rich datasets can close the gap on frontier labs in narrow but valuable domains. Results so far suggest the answer leans yes for certain tasks. Thomson tops some composites on legal hardness and long-context handling. It admits when it doesn't know. It cites sources. These traits matter more to a partner at a law firm than raw benchmark scores. The road ahead involves wider rollout. More features in CoCounsel. Potential expansion to tax and regulatory products. Continued benchmarking transparency. And ongoing collaboration with Anthropic even as internal capabilities grow. Thomson Reuters has placed its chips. The house it builds won't replace every rented roof. But over time that equity could compound into a meaningful lead in professional AI. Lawyers and compliance officers will decide if the bet pays off. Their verdict will shape the next wave of enterprise AI strategy.
Amazon's $220 Billion AI Bet: How an Anthropic Partnership Powers Cloud Dominance
Amazon raised its capital spending target for next year to $220 billion. The increase of $20 billion stems largely from higher memory costs. Yet the move signals far more than inflation in components. It reflects a calculated wager on artificial intelligence infrastructure that has already begun to pay off in both revenue acceleration and paper gains from a marquee investment. The announcement came tucked inside the company's second-quarter earnings release. Revenue climbed 20 percent from a year earlier. AWS posted its fastest growth in more than four years. Operating income jumped 43 percent to $27.5 billion, with the cloud unit contributing the bulk at $16.6 billion. The Motley Fool laid out the numbers in detail the same day. But the real story sits off the operating income line. Amazon's stake in Anthropic generated massive non-operating gains. In the first quarter alone the company recorded $16.8 billion in pre-tax income from the position, according to Yahoo Finance. By the second quarter that figure had swelled further, pushing total non-operating pre-tax other income to $53.4 billion, Business Insider reported. The partnership began years earlier. Amazon first put $4 billion into the AI startup in 2023, followed by another $4 billion in late 2024. Then came the big expansion. In April 2026 the two sides agreed on an additional $5 billion immediately and up to $20 billion more tied to commercial milestones. Total committed capital reached $13 billion so far with room to climb. Bloomberg broke the terms the day they were announced. Anthropic gave something back. The startup pledged to spend more than $100 billion on AWS services over the next decade. That commitment secures up to five gigawatts of new computing capacity built around Amazon's custom Trainium and Graviton chips. "Our custom AI silicon offers high performance at significantly lower cost for customers, which is why it's in such hot demand," Amazon CEO Andy Jassy said in the official release from AboutAmazon.com. The arrangement locks in demand. It also raises the bar for everyone else. Amazon, Microsoft and Alphabet together control more than 60 percent of the cloud market. Oracle sits a distant fourth at roughly four percent. Those scale advantages compound when hyperscalers pour tens of billions into data centers that smaller players cannot match. Investors have taken notice. Amazon shares rose more than 10 percent year to date by late August and have outperformed the S&P 500. The Anthropic position alone, once carried at cost, now reflects valuations that imply a stake worth between $180 billion and $240 billion at recent marks. An eventual IPO for Anthropic, confidentially filed in June and eyed for as early as October, could crystallize even larger gains. TechCrunch captured the mutual benefits when the April deal closed. Yet the spending surge carries risks. Free cash flow collapsed in the first quarter as capital expenditures hit $44.2 billion. Memory prices have climbed. AWS must keep delivering growth fast enough to offset the outlays. So far it has. AI and chips businesses each crossed $25 billion in annualized revenue run rate. Backlog sits at $496 billion with triple-digit growth in key segments. Andy Jassy has spoken of AWS as a potential trillion-dollar revenue business over time. That target looks less fanciful when a single partner like Anthropic commits to $100 billion in spend and multiple other AI labs sign multi-gigawatt deals. "The fact that we have multi-year, multi-gigawatt commitments from the two largest AI labs in Anthropic and OpenAI, and more and more companies using Trainium is exciting and promising," Jassy noted during the earnings call, as quoted by Variety. Wall Street's reaction mixed optimism with caution. Some analysts point to the mark-to-market accounting that inflated recent profits. Strip out the $16.8 billion paper gain from the first quarter and operating results still looked strong, but the distinction matters for valuation. The Next Web highlighted exactly that tension in May. Amazon's approach differs from pure-play AI labs that burn cash without the offsetting cloud revenue. The company can pass higher costs to customers who need more powerful instances as their models scale. It can also amortize the infrastructure spend across a broad base of enterprise workloads that extend beyond generative AI. Recent X conversations reflect the same debate. One investor noted that "Amazon's Anthropic stake came with cloud compute commitments, not just cash for equity. The 2T valuation bump looks good on paper, but AMZN's real upside is AWS consumption tied to that partnership." Others flagged Anthropic's $65 billion revenue run rate as a tailwind for Amazon, Google and Nvidia. The competitive moat widens with every gigawatt added. New entrants face years of lead time and billions in upfront capital before they can offer comparable performance at competitive prices. Amazon's Trainium chips already deliver meaningful cost advantages. Expanding that advantage while locking in the largest AI developers creates a self-reinforcing cycle. Of course execution still counts. Supply chains for memory and power remain tight. Regulatory scrutiny of big tech infrastructure builds could intensify. And valuations for private AI companies have climbed fast. Anthropic's implied worth topped $965 billion in its latest round and secondary trades pushed higher still. Even so, the numbers paint a picture of a company turning massive spending into measurable returns. Revenue growth outpacing capex growth. Cloud operating margins expanding. A strategic investment that has already delivered billions in recognized gains and promises more. Jeff Bezos no longer runs day-to-day operations, but his successor's willingness to spend at this scale keeps the original vision of infrastructure dominance alive. Whether the $220 billion bet proves conservative or not will show in the quarters ahead. Demand forecasts already stretch into 2027 and 2028. If AWS continues to accelerate, that number may rise again. For now the market seems willing to underwrite the outlay. Amazon's shares reflect confidence that the infrastructure built today will anchor AI workloads for years to come.
CFTC, US soldier Accused of Illegal Polymarket Bet Spar over Interpretation of Prediction Markets
A judge stayed the CFTC's civil case against a soldier who allegedly used nonpublic information for a Polymarket bet, but the regulator is trying to weigh in on the criminal case. Gannon Ken Van Dyke, a soldier who allegedly made more than $400,000 using nonpublic information to trade event contracts on prediction market platform Polymarket, is pushing back against attempts by the US Commodity Futures Trading Commission (CFTC) to weigh in on his criminal case. In a Monday filing in the US District Court for the Southern District of New York, Van Dyke's lawyers opposed the CFTC's efforts to file an amicus brief relating to claims in his case. The letter referred to the commodities regulator's counsel filing a notice requesting permission from the court to provide its views on many of Van Dyke's defense claims, including that event contracts on platforms like Polymarket were not "swaps" under the purview of the CFTC. "The CFTC is no sheep 'friend of the Court' here," said defense attorneys. "It is a regulatory wolf, with its own case against Mr. Van Dyke that it refuses to pursue itself. Rather, like a true coursing predator, the CTFC seeks to advance its own interests through the back door of an amicus brief instead of facing its own case against Mr. Van Dyke head on. This Court should not countenance the CFTC's litigation gambits." US authorities charged Van Dyke with fraud in April over allegations he traded event contracts on Polymarket related to the removal of Venezuelan President Nicolás Maduro in January, an operation for which he was privy to nonpublic information. The incident has been one of the leading cases lawmakers and critics of prediction markets point to as an example of potential manipulation on platforms like Kalshi and Polymarket. A federal judge already ordered that the CFTC's civil case against Van Dyke be stayed "pending the outcome of the criminal proceeding." He has pleaded not guilty to all charges and a criminal trial could potentially begin in late 2026 or early 2027.
While GTA 6 leaker runs rampant, Discord says it hasn't been served a subpoena yet
The Grand Theft Auto 6 leaker is still at large, and Take-Two Interactive's attempts to track them down have thus far not worked out. But the story, as usual, continues. Today the latest update came from Discord's marketing director Ryan Rigney, who said that the company ahs not yet been served a subpoena, despite an Aug. 20 court filing saying that the publisher was looking for help from the chat app and Microsoft alike. Last week's court filing said it would be asking Discord for "all user accounts that are/were members communicating with the Discord server(s)" that involve "CYBERLEEK," "CINEMATICROCKSTAR," and "Surfer24k™ (+ replacement handles cyberleek_west, surwest)," as well as "Ødyssey.gg," "'! Odyssey' public brand guild (discord.gg/odyssey)," and even YouTuber/Twitch streamer "DarkViperAU editors' server (discord.gg/darkviperau)." "Discord has not yet been served with a subpoena from Take Two," Rigney said today on Twitter/X. "When we do, we'll evaluate the validity and scope before responding." The subpoenas have raised some concerns that unknowing parties could get caught in the crossfire simply for engaging with the leaker, who has been drip-feeding short video clips of the game in action for the past week ahead of its full reveal on Netflix this Thursday. They have also been heavily promoting a cryptocurrency throughout the ordeal. The same court filings have called on Microsoft to give up "all internal Microsoft business records and investigative records associated with Microsoft's internal investigation of the 'cyberleek' persona sufficient to identify the user(s), person(s) and/or entity/ies associated with that persona." The GTA 6 leaker continued today to post new video clips along with allowing people to vote on what they will leak next. Today's video showcased main character Jason at a nudist colony somewhere in Vice City. With just days to go before the game's reveal, the hunt is still on for the leaker who will likely face charges if and when they are caught.
