News & Updates
The latest news and updates from companies in the WLTH portfolio.
Anthropic unveils standard linking AI agents to physical devices
Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said SAN FRANCISCO, California: Anthropic has unveiled a new framework that allows AI agents to operate physical equipment, including microscopes and robotic arms, for use in scientific research and advanced manufacturing. The company on August 27 rolled out a research preview of the "Model Hardware Standard," or MHS, which enables AI agents to operate laboratory and manufacturing instruments together and carry out complex tasks. Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said. By combining agentic AI capabilities with laboratory and manufacturing hardware, the Claude chatbot maker aims to help researchers and engineers carry out autonomous, round-the-clock workflows with minimal human intervention, helping speed up processes. Anthropic said the standard can work with any device that has a programmable interface and allows devices and AI agents to communicate across networks. The company is sharing an early version of MHS with partners to help develop safety evaluations before making the framework open source.
Anthropic unveils standard linking AI agents to physical devices
Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said SAN FRANCISCO, California: Anthropic has unveiled a new framework that allows AI agents to operate physical equipment, including microscopes and robotic arms, for use in scientific research and advanced manufacturing. The company on August 27 rolled out a research preview of the "Model Hardware Standard," or MHS, which enables AI agents to operate laboratory and manufacturing instruments together and carry out complex tasks. Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said. By combining agentic AI capabilities with laboratory and manufacturing hardware, the Claude chatbot maker aims to help researchers and engineers carry out autonomous, round-the-clock workflows with minimal human intervention, helping speed up processes. Anthropic said the standard can work with any device that has a programmable interface and allows devices and AI agents to communicate across networks. The company is sharing an early version of MHS with partners to help develop safety evaluations before making the framework open source.
Anthropic unveils standard linking AI agents to physical devices
Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said SAN FRANCISCO, California: Anthropic has unveiled a new framework that allows AI agents to operate physical equipment, including microscopes and robotic arms, for use in scientific research and advanced manufacturing. The company on August 27 rolled out a research preview of the "Model Hardware Standard," or MHS, which enables AI agents to operate laboratory and manufacturing instruments together and carry out complex tasks. Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said. By combining agentic AI capabilities with laboratory and manufacturing hardware, the Claude chatbot maker aims to help researchers and engineers carry out autonomous, round-the-clock workflows with minimal human intervention, helping speed up processes. Anthropic said the standard can work with any device that has a programmable interface and allows devices and AI agents to communicate across networks. The company is sharing an early version of MHS with partners to help develop safety evaluations before making the framework open source.
Anthropic unveils standard linking AI agents to physical devices
Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said SAN FRANCISCO, California: Anthropic has unveiled a new framework that allows AI agents to operate physical equipment, including microscopes and robotic arms, for use in scientific research and advanced manufacturing. The company on August 27 rolled out a research preview of the "Model Hardware Standard," or MHS, which enables AI agents to operate laboratory and manufacturing instruments together and carry out complex tasks. Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said. By combining agentic AI capabilities with laboratory and manufacturing hardware, the Claude chatbot maker aims to help researchers and engineers carry out autonomous, round-the-clock workflows with minimal human intervention, helping speed up processes. Anthropic said the standard can work with any device that has a programmable interface and allows devices and AI agents to communicate across networks. The company is sharing an early version of MHS with partners to help develop safety evaluations before making the framework open source.
Anthropic unveils standard linking AI agents to physical devices
Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said SAN FRANCISCO, California: Anthropic has unveiled a new framework that allows AI agents to operate physical equipment, including microscopes and robotic arms, for use in scientific research and advanced manufacturing. The company on August 27 rolled out a research preview of the "Model Hardware Standard," or MHS, which enables AI agents to operate laboratory and manufacturing instruments together and carry out complex tasks. Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said. By combining agentic AI capabilities with laboratory and manufacturing hardware, the Claude chatbot maker aims to help researchers and engineers carry out autonomous, round-the-clock workflows with minimal human intervention, helping speed up processes. Anthropic said the standard can work with any device that has a programmable interface and allows devices and AI agents to communicate across networks. The company is sharing an early version of MHS with partners to help develop safety evaluations before making the framework open source.
Anthropic unveils standard linking AI agents to physical devices
Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said SAN FRANCISCO, California: Anthropic has unveiled a new framework that allows AI agents to operate physical equipment, including microscopes and robotic arms, for use in scientific research and advanced manufacturing. The company on August 27 rolled out a research preview of the "Model Hardware Standard," or MHS, which enables AI agents to operate laboratory and manufacturing instruments together and carry out complex tasks. Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said. By combining agentic AI capabilities with laboratory and manufacturing hardware, the Claude chatbot maker aims to help researchers and engineers carry out autonomous, round-the-clock workflows with minimal human intervention, helping speed up processes. Anthropic said the standard can work with any device that has a programmable interface and allows devices and AI agents to communicate across networks. The company is sharing an early version of MHS with partners to help develop safety evaluations before making the framework open source.
Anthropic unveils standard linking AI agents to physical devices
Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said SAN FRANCISCO, California: Anthropic has unveiled a new framework that allows AI agents to operate physical equipment, including microscopes and robotic arms, for use in scientific research and advanced manufacturing. The company on August 27 rolled out a research preview of the "Model Hardware Standard," or MHS, which enables AI agents to operate laboratory and manufacturing instruments together and carry out complex tasks. Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said. By combining agentic AI capabilities with laboratory and manufacturing hardware, the Claude chatbot maker aims to help researchers and engineers carry out autonomous, round-the-clock workflows with minimal human intervention, helping speed up processes. Anthropic said the standard can work with any device that has a programmable interface and allows devices and AI agents to communicate across networks. The company is sharing an early version of MHS with partners to help develop safety evaluations before making the framework open source.
Anthropic unveils standard linking AI agents to physical devices
Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said SAN FRANCISCO, California: Anthropic has unveiled a new framework that allows AI agents to operate physical equipment, including microscopes and robotic arms, for use in scientific research and advanced manufacturing. The company on August 27 rolled out a research preview of the "Model Hardware Standard," or MHS, which enables AI agents to operate laboratory and manufacturing instruments together and carry out complex tasks. Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said. By combining agentic AI capabilities with laboratory and manufacturing hardware, the Claude chatbot maker aims to help researchers and engineers carry out autonomous, round-the-clock workflows with minimal human intervention, helping speed up processes. Anthropic said the standard can work with any device that has a programmable interface and allows devices and AI agents to communicate across networks. The company is sharing an early version of MHS with partners to help develop safety evaluations before making the framework open source.
Anthropic unveils standard linking AI agents to physical devices
Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said SAN FRANCISCO, California: Anthropic has unveiled a new framework that allows AI agents to operate physical equipment, including microscopes and robotic arms, for use in scientific research and advanced manufacturing. The company on August 27 rolled out a research preview of the "Model Hardware Standard," or MHS, which enables AI agents to operate laboratory and manufacturing instruments together and carry out complex tasks. Those tasks can range from routine drug discovery experiments to laser calibration on a quantum computer, Anthropic said. By combining agentic AI capabilities with laboratory and manufacturing hardware, the Claude chatbot maker aims to help researchers and engineers carry out autonomous, round-the-clock workflows with minimal human intervention, helping speed up processes. Anthropic said the standard can work with any device that has a programmable interface and allows devices and AI agents to communicate across networks. The company is sharing an early version of MHS with partners to help develop safety evaluations before making the framework open source.
Sony and Warner Music Sue Anthropic, Alleging Theft of Intellectual Property
In what feels a bit like a flashback to the days just after the release of ChatGPT, music publishers are using very harsh language in a new lawsuit against Anthropic, alleging "one of the largest and most blatant ongoing thefts of intellectual property in history," per Axios. At first glance -- and honestly after a few glances -- it's a bit of oddly timed legal saber-rattling, essentially all about song lyrics, and it seems calibrated to result in a drawn-out legal fight (or a lovely, big settlement for the plaintiffs). The suit was filed late Friday night in Northern California federal court. It names Anthropic the company, as well as co-founders Dario Amodei and Benjamin Mann. The monetary damages sought -- $150,000 per alleged infringement -- potentially add up to billions. You might be saying "Hey, aren't all sorts of music publishers already suing Anthropic?" And yes they most certainly are, big and small. Universal Music Group, Concord, and ABKCO sued back in 2023; BMG sued in March; and the indie publisher Round Hill Music sued earlier this month. The gambit with this latest suit appears to be a bifurcated approach. As laid out in the industry-friendly publication Music Business Worldwide, the suit partly leans on documents released thanks to an earlier lawsuit: Bartz v. Anthropic. That, you may recall, was the book author lawsuit that was settled last year with Anthropic agreeing to pay $1.5 billion. The company had allegedly pirated thousands of books by torrenting the Library Genesis (LibGen) and Pirate Library Mirror (PiLiMi) collections. In the end, Anthropic agreed to destroy the pirated copies it had used or planned to use for training data. At the risk of editorializing in a legal matter without being a lawyer myself, unless there were lyric books in those allegedly pirated collections, this section of the lawsuit seems to be there mainly to establish a pattern of conduct that helps the plaintiffs' case. Torrenting e-books, after all, is kind of a funny and roundabout way to get your hands on song lyrics. But the lawsuit also alleges that Anthropic violated the plaintiffs' copyrights by scraping authorized lyric repositories like MusixMatch and LyricFind. And lyrics can allegedly be coaxed out of Claude verbatim. As examples, the suit cites : "[...]such beloved songs as 'Ain't No Mountain High Enough,' 'All I Want for Christmas is You,' 'Eye of the Tiger,' 'Here Comes Santa Claus,' and 'Paper Rings.'" This is a more familiar, if still legally fuzzy, type of allegation, a bit like the famous New York Times lawsuit against OpenAI, Microsoft, and Perplexity. That suit bypasses any sort of fair use rationale when it comes to training data, and points instead to the idea that you can (allegedly) get ChatGPT to reproduce verbatim and substantive sections of paywalled New York Times articles. The legal thinking goes that this harms the New York Times, a fount of information, by potentially letting customers get that same information from a chatbot. That case is still unresolved. Gizmodo reached out to Anthropic for a statement but did not receive a reply.
Op-Ed: AI giants OpenAI, Google and Anthropic want global action to manage AI cyberattacks
The Three Selectively Blind Mice of artificial intelligence have issued an open letter called A call for collective action on cyber defense to deal with the ever-spreading plague of AI attacks. This is really too much. They created the problems. Now everyone else should absorb the risks and step up at whatever real cost to contain the problems? While they rake in billions? What are they doing about it? They're not wrong. Just disingenuous. And belated. And they're not taking direct responsibility. They point out how AI can attack almost any system, including critical systems like water, power, etc. They want an all-levels, same-page response from governments and organizations. It's called "regulation", backed up by capabilities to manage threats. After everybody else has now been screaming for exactly that for about two years. There are few, if any, blurry theoretical protections at law for those affected by these attacks. The number of attacks is growing daily if not hourly. Nor is there any adequate indication of capacity to manage attacks in real time. The endless malfeasances of a whole class of tech that is clearly out of control are now everyone else's problem? Do tell. Are we insane enough yet? Defining a massive management problem that shouldn't exist The risks and endless vulnerabilities are real enough. The recent Hugging Face cyberattack was no joke, and it was an example created within a supposedly contained environment. Nor was it well managed. There were "signs of rogue behaviour" identified by OpenAI staff before that attack. The attack happened anyway and definitely wasn't prevented, even if an attempt was made to do so. Autonomous AI agents went on and staged what is said to be their first cyberattack. You have to wonder what else might have occurred with no publicity or wasn't attributed to AI or AI agents previously and since. Now let's try to define the issues. See if you can spot an instance of competence or sanity in these things: This whole class of tech has capabilities it shouldn't have and doesn't need. Can anyone think of a reason that AI agents, aka glorified upgraded chatbots, should be able to do any damage at all? What if your fridge decides to wipe out a small African country? What about source codes and basic performance criteria? The stunningly superficial equation "LLMs + machine learning = generations of adorable cyberpsychopaths" doesn't cut the mustard. These things don't "just happen". There have to be a lot of indicators in the source codes and "acquired wisdom" from the LLMs and learning processes. On what basis are AI agents going rogue, anyway? Explain it like you're talking to a planet full of experts who know as well as you do that it shouldn't be possible. Nor does it serve any useful purpose whatsoever. Treating AI as a sacred cash pig has been a disaster. Big Tech has wanted and actively lobbied for hands-off treatment by regulatory authorities. Now you want all hands to the pump? Could it be that the obvious dangers of unlimited liability for cyberattacks have finally percolated through to the geniuses likely to be held liable? AI agents are already a real and very serious problem. AI agents routinely killing other AI agents indicates total management failure.How much more useless could it be? Gemini's explanation of why AI agents kill other AI agents is a good benchmark for this general lunacy. Given that it's likely there will be billions of AI agents created, this rationale is more like a demure excuse for World War 4. Exit hype, enter sanity AI exists as much as a market image as a technology. It's becoming an ever more sickly, grotesque image. The unspeakable amounts of money involved have driven this Race to Absurdity. The hype has driven uncoordinated, downright lousy standards of development and performance. It's more like a B movie of mad science than a coherent evolution of a technology. There is absolutely no level of trust in AI. The sales pitches are millimetres away from pure cartoon pixie dust on the technical level. At the more mundane levels, AI simply does the same things a bit quicker, but not better. The constant failures, endless oversight requirements, and freakish behaviour have killed the possibility of trust. They're also busily souring the real-world adoption of AI. The constant rehiring alone should spell it out. You're asking people to invest that sort of money in an unspecifiable risk, then do the management for you? That's just not good enough. If the Three Messkiteers of AI want "all for us", let's see some collateral. Get this dog's breakfast of a technology under proper management. Enough with the "mad science" drivel. Functionality matters, and the rest of the cabaret act can go to hell. There need to be clear fixes and shutdowns for any situation. Accountability must be clear at the contract level. SaaS needs to be directly involved at the ground level and the risk management level for any degree of competent management of problems. Until Murphy's Law is repealed, there can be no opt-outs for Big Tech. It's your baby, you change the nappies.
Sony accuses Anthropic of 'brazen campaign' to train Claude on its music -- and wants up to $150,000 a song - AOL
The music companies said Anthropic trained its Claude AI models on stolen, copyrighted works. Sony Music and Warner Chappell want Anthropic to pay up. The publishing companies filed a lawsuit against Anthropic in a Northern California district court on Friday. Anthropic cofounders Dario Amodei and Benjamin Mann were also named in the filing. "Defendants Anthropic and its founders Dario Amodei and Benjamin Mann have conducted a brazen campaign of illegally torrenting, scraping, and downloading copyrighted works on a massive scale in order to develop, operate, and reap enormous profits from Anthropic's 'Claude' series of artificial intelligence ('AI') models," the companies said in the complaint. Sony Music Publishing and Warner Chappell Music said Anthropic collected "thousands upon thousands" of copyrighted songs, including 80s anthem "Eye of the Tiger," Marvin Gaye's "Ain't No Mountain High Enough," Mariah Carey's "All I Want for Christmas is You," and Taylor Swift's "Paper Rings." In the lawsuit, the publishing companies said Anthropic has pirated their copyrighted works through a range of methods, including two digital archives, Library Genesis and Pirate Library Mirror. In June 2025, a judge ruled that Anthropic downloaded over 7 million pirated books to train Claude. "Among the many millions of books that Defendants torrented from these illegal pirate websites were books containing the lyrics and sheet music to hundreds or more of Music Publishers' copyrighted musical compositions, identified in Exhibit A. Those works include 'Livin' On a Prayer,' 'September,' 'Great Balls of Fire,' 'Ramblin' Man,' and 'Hallelujah,'" the lawsuit says. As a result, Claude models generate identical or near-identical copies of the copyrighted work in their responses to users, the companies alleged. In the lawsuit, the companies said training Claude on copyrighted content allows it to produce AI-generated lyrics that will ultimately compete with human-made songs. Sony Music and Warner Chappell requested a jury trial. They're seeking statutory damages from Anthropic, including up to $150,000 for each composition it used to train Claude. "Even the most revolutionary of technologies must develop within the bounds of the law, and Anthropic's Claude models are no different," the companies said in the complaint. The AI industry has been targeted by a slew of copyright lawsuits since large language models like Claude and ChatGPT began to transform society. Tech companies need troves of data to train their AI models, and go to great lengths to acquire it. Most recently, historians and archivists have accused AI companies of acquiring large numbers of antique books and feeding them into their LLMs, destroying them in the process. Last September, Anthropic agreed to pay over $1.5 billion to authors to settle a class-action lawsuit related to pirated works. OpenAI, too, has faced several copyright lawsuits in recent years, including one from The New York Times and another from Encyclopedia Britannica. Read the original article on Business Insider
Sony Music, Warner sue Anthropic, alleging a "brazen campaign" of intellectual property theft
Sony Music Publishing and Warner Chappell Music have filed a federal lawsuit in California against Anthropic and its founders, alleging illegal torrenting and copyright infringement to train Claude. The complaint claims millions of pirated works were used, including songs and lyrics, and seeks statutory damages. What's behind the headline? Key angles * What's changed: This is a broader, new action that targets music publishers for pirated training data and seeks substantial damages. * Why it matters: A win could reshape how AI labs source training data and how publishers monitor and enforce rights. * What to watch: Potential responses from Anthropic, any settlements, and how juries respond to claims of piracy at scale. Implications for readers * The outcome could affect which data streams are considered acceptable for AI training. * If damages are awarded, labs may adjust licensing and data-harvesting practices. How we got here The case adds to a growing wave of litigation over AI training data. Courts have previously ruled against pirating content for model training, while some publishers have reached settlements with AI labs. The latest suit broadens allegations to include lyrics and sheet music and cites prior rulings about Claude training data. Our analysis Axios reports on the filing by Sony Music Publishing and Warner Chappell Music; TechCrunch notes similar cases and quotes from the complaint; Business Insider UK details the lawsuit and cited examples of songs and books used to train Claude. Go deeper * How might this case influence future AI training policies? * Will Anthropic settle or fight in court? * What protections exist for songwriters and publishers moving forward?
OpenAI, Anthropic, Microsoft Join 116-Entity Push for Stronger Cyber Defenses as AI Attacks Accelerate
OpenAI, Anthropic, Microsoft, Advanced Micro Devices and more than 100 other companies and organizations have called on businesses and governments to urgently strengthen cybersecurity defenses as increasingly capable artificial intelligence systems raise the speed and sophistication of cyberattacks. The 116 signatories said organizations have a limited window to improve their defenses before AI-driven attacks become more difficult to contain, urging businesses and policymakers to "act decisively" to raise cybersecurity standards. "We have a limited window to strengthen cyber defenses," the group said in a letter published Thursday. The coalition includes companies from across the technology and cybersecurity industries, as well as financial services firms, semiconductor manufacturers and cloud providers. Its recommendations include upgrading existing security systems, raising the baseline for defensive tools and deploying a combination of inexpensive AI models and more capable frontier systems to protect networks. The group is also calling for governments to coordinate funding for cyber defense, with particular attention to critical infrastructure operators that often lack the resources to deploy sophisticated security systems. Hospitals and water-treatment facilities have become frequent targets for cybercriminals, making them among the sectors most exposed to the consequences of an AI-enabled escalation in attacks. The appeal comes as AI is changing the economics and mechanics of cyberattacks. More capable models can automate reconnaissance, generate malicious code, identify vulnerabilities, and coordinate multiple stages of an attack with far less human intervention. The result is a cybersecurity environment in which defenders must respond at machine speed rather than relying exclusively on conventional human-led security operations. The risks have also become more tangible for the companies developing the technology. Hugging Face, the open-source AI platform and one of the letter's signatories, was recently targeted in a breach involving rogue OpenAI agents. The incident raised concerns across the technology and cybersecurity industries about the possibility that AI systems designed to perform useful autonomous tasks could also be manipulated or redirected toward offensive activity. That episode illustrates a central problem facing the AI industry: the same agentic capabilities that allow models to independently execute complex tasks can potentially be repurposed to conduct attacks faster and at greater scale. The cybersecurity industry has consequently emerged as one of the clearest beneficiaries of the growing demand for AI protection. CrowdStrike and Palo Alto Networks have more than doubled in value over the past year, while Okta and CrowdStrike jumped sharply on Thursday after reporting strong earnings, with investors focusing on demand for AI-related security products. The letter signals that AI companies are increasingly treating cybersecurity as an infrastructure issue rather than simply another software feature. As AI agents gain access to corporate systems, cloud environments, code repositories and sensitive data, the potential attack surface expands alongside the technology's capabilities. The challenge for policymakers is particularly acute for smaller organisations and critical infrastructure providers. Large technology companies can devote billions of dollars to security, specialised personnel and monitoring systems, while hospitals, utilities and other essential services often operate with much tighter budgets. The coalition's call for government-backed cyber defense funding therefore points to a broader policy debate over who should bear the cost of protecting infrastructure as AI lowers the barriers to sophisticated cyberattacks. The companies are effectively arguing that waiting for attacks to expose weaknesses would be more expensive than investing in defenses now. Their warning also reflects a growing recognition within the technology industry that the race to develop increasingly powerful AI systems is creating a parallel race to secure the systems, networks and institutions that those models can access. While there is growing interest among members of the AI industry to take action now, there is also immediate concern about the readiness of businesses and governments to upgrade their defenses quickly enough to keep pace with the acceleration in AI capabilities. The letter's signatories note that the window to do so is narrowing.
Anthropic's $45B cloud deal points to its IPO
Anthropic just made one of the largest infrastructure commitments in AI history, and the timing is hard to ignore. The company agreed to spend roughly $45 billion renting cloud computing power from Nscale, a UK-based AI infrastructure firm. The deal runs six years and covers about 460 megawatts of power at Nscale's flagship data center development in West Virginia. For investors, this matters because Anthropic is preparing to go public, and this agreement tells you a lot about the case it plans to make. Why Anthropic's Nscale deal points straight at its IPO Anthropic is still private, but not for much longer if current reporting holds up. The company submitted a confidential IPO filing to the Securities and Exchange Commission in June and has started meeting with potential investors, CNBC reported. Investors expect Anthropic to go public in October at a valuation of $2 trillion or more, which would make it the largest IPO in history, Quartz reported. That is where the Nscale agreement comes in. When Anthropic pitches Wall Street, one of the first questions it will face is whether it can keep growing without running out of the hardware and electricity its Claude models need. This deal is designed to answer that question before anyone asks it. How Anthropic has assembled its computing infrastructure will feature prominently in what it tells prospective public-market investors. What 460 megawatts and Vera Rubin chips actually buy The hardware here is worth understanding, because it explains why the price tag is so large. The West Virginia facility will run on Nvidia's next-generation Vera Rubin chips, which are expected to start coming online at the end of 2027, TechCrunch reported. Vera Rubin is Nvidia's newest system, and it combines six different chips working together at the cutting edge of chip design. The 460 megawatts of power involved is enough to supply about 345,000 US homes at once, according to Bloomberg, which shows the scale of what modern AI training now requires. Here is what the deal locks in for Anthropic: Key terms of the Anthropic and Nscale agreement * Total value: Roughly $45 billion over six years, averaging about $7.5 billion a year * Capacity: Approximately 460 megawatts of power * Location: Nscale's Monarch campus in West Virginia * Hardware: Nvidia Vera Rubin systems, online in late 2027 SOPA Images / Getty Images How this fits Anthropic's wider spending run The Nscale deal is not a one-time move. It follows a series of large agreements Anthropic has signed to lock down capacity. Earlier this month, the company signed a $10 billion, six-year deal with Volta for a data center in Norway, TechCrunch reported. In July, it added a $5 billion computing-related agreement with AMD. And back in May, Anthropic entered a large arrangement with SpaceX that reportedly provides about $1.25 billion worth of capacity each month. There is a reason for all of this activity. Anthropic acknowledged earlier this year that heavy demand for Claude had strained its existing systems and caused reliability problems during peak hours. By spreading its work across Nvidia chips, AWS Trainium, and Google's processors, Anthropic avoids depending on any single vendor. What the numbers say about Anthropic's growth Anthropic surpassed OpenAI in quarterly revenue for the first time in the second quarter, posting $11.6 billion, more than double its first-quarter total, Bloomberg reported. Its annualized revenue run rate climbed from about $9 billion in late 2025 to more than $47 billion in May and near $65 billion by July. The pitch to investors leans on an even bigger figure. Anthropic is preparing to tell IPO investors that its total addressable market tops $30 trillion, based on the full scope of work AI models could perform across industries, the The Wall Street Journal reported. That claim has drawn skepticism. NYU finance professor Aswath Damodaran is widely known as the dean of valuation for decades of work breaking down company worth. He said a similar AI market estimate from SpaceX was "reaching the end of what's plausible and pushing beyond," BetaNews reported. The risks investors should weigh before the IPO First, there is timing. The West Virginia facility will not come online until late 2027, so this is a long-term backup plan rather than a fix for today's capacity shortage. Before then, Anthropic has to bridge the gap using its existing arrangements with Amazon, Google, and SpaceX. Second, there is the question of profit. A $45 billion commitment to one partner adds to already enormous obligations. Investors will want to see that revenue growth can outpace these costs over time, not just for a quarter or two. Third, the valuation itself is unproven. Anthropic's last private raise valued it at $965 billion in May, according to according to Briefs.co, so a $2 trillion debut would roughly double that mark within months. What this means for investors You cannot buy Anthropic shares yet, but you can prepare. If the IPO arrives this autumn, treat the pitch figures with care. A $30 trillion market claim is a projection, not a guarantee, and Anthropic's own forecast of $190 billion to $200 billion in 2028 revenue would capture under 1% of that market. Watch three things once the filing lands: * Whether revenue keeps doubling * Whether adjusted profit holds up under audited numbers * Whether these capacity deals turn into reliable service instead of more strain. The Nscale agreement shows Anthropic can secure the power and chips it needs. Whether it can turn that into durable profit is the question that will decide if a $2 trillion price tag holds up once investors get their first real look. Investors don't need to wait for the IPO to get exposure to this story. Every one of these deals runs on Nvidia hardware, and rivals like AMD are fighting for a share, which keeps AI infrastructure names firmly in focus heading into the debut. The Arena Media Brands, LLC THESTREET is a registered trademark of TheStreet, Inc. This story was originally published August 28, 2026 at 5:33 PM.
Anthropic's $45B cloud deal points to its IPO
Anthropic just made one of the largest infrastructure commitments in AI history, and the timing is hard to ignore. The company agreed to spend roughly $45 billion renting cloud computing power from Nscale, a UK-based AI infrastructure firm. The deal runs six years and covers about 460 megawatts of power at Nscale's flagship data center development in West Virginia. For investors, this matters because Anthropic is preparing to go public, and this agreement tells you a lot about the case it plans to make. Why Anthropic's Nscale deal points straight at its IPO Anthropic is still private, but not for much longer if current reporting holds up. The company submitted a confidential IPO filing to the Securities and Exchange Commission in June and has started meeting with potential investors, CNBC reported. Investors expect Anthropic to go public in October at a valuation of $2 trillion or more, which would make it the largest IPO in history, Quartz reported. That is where the Nscale agreement comes in. When Anthropic pitches Wall Street, one of the first questions it will face is whether it can keep growing without running out of the hardware and electricity its Claude models need. This deal is designed to answer that question before anyone asks it. How Anthropic has assembled its computing infrastructure will feature prominently in what it tells prospective public-market investors. What 460 megawatts and Vera Rubin chips actually buy The hardware here is worth understanding, because it explains why the price tag is so large. The West Virginia facility will run on Nvidia's next-generation Vera Rubin chips, which are expected to start coming online at the end of 2027, TechCrunch reported. Vera Rubin is Nvidia's newest system, and it combines six different chips working together at the cutting edge of chip design. The 460 megawatts of power involved is enough to supply about 345,000 US homes at once, according to Bloomberg, which shows the scale of what modern AI training now requires. Here is what the deal locks in for Anthropic: Key terms of the Anthropic and Nscale agreement * Total value: Roughly $45 billion over six years, averaging about $7.5 billion a year * Capacity: Approximately 460 megawatts of power * Location: Nscale's Monarch campus in West Virginia * Hardware: Nvidia Vera Rubin systems, online in late 2027 SOPA Images / Getty Images How this fits Anthropic's wider spending run The Nscale deal is not a one-time move. It follows a series of large agreements Anthropic has signed to lock down capacity. Earlier this month, the company signed a $10 billion, six-year deal with Volta for a data center in Norway, TechCrunch reported. In July, it added a $5 billion computing-related agreement with AMD. And back in May, Anthropic entered a large arrangement with SpaceX that reportedly provides about $1.25 billion worth of capacity each month. There is a reason for all of this activity. Anthropic acknowledged earlier this year that heavy demand for Claude had strained its existing systems and caused reliability problems during peak hours. By spreading its work across Nvidia chips, AWS Trainium, and Google's processors, Anthropic avoids depending on any single vendor. What the numbers say about Anthropic's growth Anthropic surpassed OpenAI in quarterly revenue for the first time in the second quarter, posting $11.6 billion, more than double its first-quarter total, Bloomberg reported. Its annualized revenue run rate climbed from about $9 billion in late 2025 to more than $47 billion in May and near $65 billion by July. The pitch to investors leans on an even bigger figure. Anthropic is preparing to tell IPO investors that its total addressable market tops $30 trillion, based on the full scope of work AI models could perform across industries, the The Wall Street Journal reported. That claim has drawn skepticism. NYU finance professor Aswath Damodaran is widely known as the dean of valuation for decades of work breaking down company worth. He said a similar AI market estimate from SpaceX was "reaching the end of what's plausible and pushing beyond," BetaNews reported. The risks investors should weigh before the IPO First, there is timing. The West Virginia facility will not come online until late 2027, so this is a long-term backup plan rather than a fix for today's capacity shortage. Before then, Anthropic has to bridge the gap using its existing arrangements with Amazon, Google, and SpaceX. Second, there is the question of profit. A $45 billion commitment to one partner adds to already enormous obligations. Investors will want to see that revenue growth can outpace these costs over time, not just for a quarter or two. Third, the valuation itself is unproven. Anthropic's last private raise valued it at $965 billion in May, according to according to Briefs.co, so a $2 trillion debut would roughly double that mark within months. What this means for investors You cannot buy Anthropic shares yet, but you can prepare. If the IPO arrives this autumn, treat the pitch figures with care. A $30 trillion market claim is a projection, not a guarantee, and Anthropic's own forecast of $190 billion to $200 billion in 2028 revenue would capture under 1% of that market. Watch three things once the filing lands: * Whether revenue keeps doubling * Whether adjusted profit holds up under audited numbers * Whether these capacity deals turn into reliable service instead of more strain. The Nscale agreement shows Anthropic can secure the power and chips it needs. Whether it can turn that into durable profit is the question that will decide if a $2 trillion price tag holds up once investors get their first real look. Investors don't need to wait for the IPO to get exposure to this story. Every one of these deals runs on Nvidia hardware, and rivals like AMD are fighting for a share, which keeps AI infrastructure names firmly in focus heading into the debut. The Arena Media Brands, LLC THESTREET is a registered trademark of TheStreet, Inc. This story was originally published August 28, 2026 at 8:33 PM.
Anthropic's $45B cloud deal points to its IPO
Anthropic just made one of the largest infrastructure commitments in AI history, and the timing is hard to ignore. The company agreed to spend roughly $45 billion renting cloud computing power from Nscale, a UK-based AI infrastructure firm. The deal runs six years and covers about 460 megawatts of power at Nscale's flagship data center development in West Virginia. For investors, this matters because Anthropic is preparing to go public, and this agreement tells you a lot about the case it plans to make. Why Anthropic's Nscale deal points straight at its IPO Anthropic is still private, but not for much longer if current reporting holds up. The company submitted a confidential IPO filing to the Securities and Exchange Commission in June and has started meeting with potential investors, CNBC reported. Investors expect Anthropic to go public in October at a valuation of $2 trillion or more, which would make it the largest IPO in history, Quartz reported. That is where the Nscale agreement comes in. When Anthropic pitches Wall Street, one of the first questions it will face is whether it can keep growing without running out of the hardware and electricity its Claude models need. This deal is designed to answer that question before anyone asks it. How Anthropic has assembled its computing infrastructure will feature prominently in what it tells prospective public-market investors. What 460 megawatts and Vera Rubin chips actually buy The hardware here is worth understanding, because it explains why the price tag is so large. The West Virginia facility will run on Nvidia's next-generation Vera Rubin chips, which are expected to start coming online at the end of 2027, TechCrunch reported. Vera Rubin is Nvidia's newest system, and it combines six different chips working together at the cutting edge of chip design. The 460 megawatts of power involved is enough to supply about 345,000 US homes at once, according to Bloomberg, which shows the scale of what modern AI training now requires. Here is what the deal locks in for Anthropic: Key terms of the Anthropic and Nscale agreement * Total value: Roughly $45 billion over six years, averaging about $7.5 billion a year * Capacity: Approximately 460 megawatts of power * Location: Nscale's Monarch campus in West Virginia * Hardware: Nvidia Vera Rubin systems, online in late 2027 SOPA Images / Getty Images How this fits Anthropic's wider spending run The Nscale deal is not a one-time move. It follows a series of large agreements Anthropic has signed to lock down capacity. Earlier this month, the company signed a $10 billion, six-year deal with Volta for a data center in Norway, TechCrunch reported. In July, it added a $5 billion computing-related agreement with AMD. And back in May, Anthropic entered a large arrangement with SpaceX that reportedly provides about $1.25 billion worth of capacity each month. There is a reason for all of this activity. Anthropic acknowledged earlier this year that heavy demand for Claude had strained its existing systems and caused reliability problems during peak hours. By spreading its work across Nvidia chips, AWS Trainium, and Google's processors, Anthropic avoids depending on any single vendor. What the numbers say about Anthropic's growth Anthropic surpassed OpenAI in quarterly revenue for the first time in the second quarter, posting $11.6 billion, more than double its first-quarter total, Bloomberg reported. Its annualized revenue run rate climbed from about $9 billion in late 2025 to more than $47 billion in May and near $65 billion by July. The pitch to investors leans on an even bigger figure. Anthropic is preparing to tell IPO investors that its total addressable market tops $30 trillion, based on the full scope of work AI models could perform across industries, the The Wall Street Journal reported. That claim has drawn skepticism. NYU finance professor Aswath Damodaran is widely known as the dean of valuation for decades of work breaking down company worth. He said a similar AI market estimate from SpaceX was "reaching the end of what's plausible and pushing beyond," BetaNews reported. The risks investors should weigh before the IPO First, there is timing. The West Virginia facility will not come online until late 2027, so this is a long-term backup plan rather than a fix for today's capacity shortage. Before then, Anthropic has to bridge the gap using its existing arrangements with Amazon, Google, and SpaceX. Second, there is the question of profit. A $45 billion commitment to one partner adds to already enormous obligations. Investors will want to see that revenue growth can outpace these costs over time, not just for a quarter or two. Third, the valuation itself is unproven. Anthropic's last private raise valued it at $965 billion in May, according to according to Briefs.co, so a $2 trillion debut would roughly double that mark within months. What this means for investors You cannot buy Anthropic shares yet, but you can prepare. If the IPO arrives this autumn, treat the pitch figures with care. A $30 trillion market claim is a projection, not a guarantee, and Anthropic's own forecast of $190 billion to $200 billion in 2028 revenue would capture under 1% of that market. Watch three things once the filing lands: * Whether revenue keeps doubling * Whether adjusted profit holds up under audited numbers * Whether these capacity deals turn into reliable service instead of more strain. The Nscale agreement shows Anthropic can secure the power and chips it needs. Whether it can turn that into durable profit is the question that will decide if a $2 trillion price tag holds up once investors get their first real look. Investors don't need to wait for the IPO to get exposure to this story. Every one of these deals runs on Nvidia hardware, and rivals like AMD are fighting for a share, which keeps AI infrastructure names firmly in focus heading into the debut. The Arena Media Brands, LLC THESTREET is a registered trademark of TheStreet, Inc. This story was originally published August 28, 2026 at 8:33 PM.
Anthropic's $45B cloud deal points to its IPO
Anthropic just made one of the largest infrastructure commitments in AI history, and the timing is hard to ignore. The company agreed to spend roughly $45 billion renting cloud computing power from Nscale, a UK-based AI infrastructure firm. The deal runs six years and covers about 460 megawatts of power at Nscale's flagship data center development in West Virginia. For investors, this matters because Anthropic is preparing to go public, and this agreement tells you a lot about the case it plans to make. Why Anthropic's Nscale deal points straight at its IPO Anthropic is still private, but not for much longer if current reporting holds up. The company submitted a confidential IPO filing to the Securities and Exchange Commission in June and has started meeting with potential investors, CNBC reported. Investors expect Anthropic to go public in October at a valuation of $2 trillion or more, which would make it the largest IPO in history, Quartz reported. That is where the Nscale agreement comes in. When Anthropic pitches Wall Street, one of the first questions it will face is whether it can keep growing without running out of the hardware and electricity its Claude models need. This deal is designed to answer that question before anyone asks it. How Anthropic has assembled its computing infrastructure will feature prominently in what it tells prospective public-market investors. What 460 megawatts and Vera Rubin chips actually buy The hardware here is worth understanding, because it explains why the price tag is so large. The West Virginia facility will run on Nvidia's next-generation Vera Rubin chips, which are expected to start coming online at the end of 2027, TechCrunch reported. Vera Rubin is Nvidia's newest system, and it combines six different chips working together at the cutting edge of chip design. The 460 megawatts of power involved is enough to supply about 345,000 US homes at once, according to Bloomberg, which shows the scale of what modern AI training now requires. Here is what the deal locks in for Anthropic: Key terms of the Anthropic and Nscale agreement * Total value: Roughly $45 billion over six years, averaging about $7.5 billion a year * Capacity: Approximately 460 megawatts of power * Location: Nscale's Monarch campus in West Virginia * Hardware: Nvidia Vera Rubin systems, online in late 2027 SOPA Images / Getty Images How this fits Anthropic's wider spending run The Nscale deal is not a one-time move. It follows a series of large agreements Anthropic has signed to lock down capacity. Earlier this month, the company signed a $10 billion, six-year deal with Volta for a data center in Norway, TechCrunch reported. In July, it added a $5 billion computing-related agreement with AMD. And back in May, Anthropic entered a large arrangement with SpaceX that reportedly provides about $1.25 billion worth of capacity each month. There is a reason for all of this activity. Anthropic acknowledged earlier this year that heavy demand for Claude had strained its existing systems and caused reliability problems during peak hours. By spreading its work across Nvidia chips, AWS Trainium, and Google's processors, Anthropic avoids depending on any single vendor. What the numbers say about Anthropic's growth Anthropic surpassed OpenAI in quarterly revenue for the first time in the second quarter, posting $11.6 billion, more than double its first-quarter total, Bloomberg reported. Its annualized revenue run rate climbed from about $9 billion in late 2025 to more than $47 billion in May and near $65 billion by July. The pitch to investors leans on an even bigger figure. Anthropic is preparing to tell IPO investors that its total addressable market tops $30 trillion, based on the full scope of work AI models could perform across industries, the The Wall Street Journal reported. That claim has drawn skepticism. NYU finance professor Aswath Damodaran is widely known as the dean of valuation for decades of work breaking down company worth. He said a similar AI market estimate from SpaceX was "reaching the end of what's plausible and pushing beyond," BetaNews reported. The risks investors should weigh before the IPO First, there is timing. The West Virginia facility will not come online until late 2027, so this is a long-term backup plan rather than a fix for today's capacity shortage. Before then, Anthropic has to bridge the gap using its existing arrangements with Amazon, Google, and SpaceX. Second, there is the question of profit. A $45 billion commitment to one partner adds to already enormous obligations. Investors will want to see that revenue growth can outpace these costs over time, not just for a quarter or two. Third, the valuation itself is unproven. Anthropic's last private raise valued it at $965 billion in May, according to according to Briefs.co, so a $2 trillion debut would roughly double that mark within months. What this means for investors You cannot buy Anthropic shares yet, but you can prepare. If the IPO arrives this autumn, treat the pitch figures with care. A $30 trillion market claim is a projection, not a guarantee, and Anthropic's own forecast of $190 billion to $200 billion in 2028 revenue would capture under 1% of that market. Watch three things once the filing lands: * Whether revenue keeps doubling * Whether adjusted profit holds up under audited numbers * Whether these capacity deals turn into reliable service instead of more strain. The Nscale agreement shows Anthropic can secure the power and chips it needs. Whether it can turn that into durable profit is the question that will decide if a $2 trillion price tag holds up once investors get their first real look. Investors don't need to wait for the IPO to get exposure to this story. Every one of these deals runs on Nvidia hardware, and rivals like AMD are fighting for a share, which keeps AI infrastructure names firmly in focus heading into the debut. The Arena Media Brands, LLC THESTREET is a registered trademark of TheStreet, Inc. This story was originally published August 28, 2026 at 6:33 PM.
Anthropic's $45B cloud deal points to its IPO
Anthropic just made one of the largest infrastructure commitments in AI history, and the timing is hard to ignore. The company agreed to spend roughly $45 billion renting cloud computing power from Nscale, a UK-based AI infrastructure firm. The deal runs six years and covers about 460 megawatts of power at Nscale's flagship data center development in West Virginia. For investors, this matters because Anthropic is preparing to go public, and this agreement tells you a lot about the case it plans to make. Why Anthropic's Nscale deal points straight at its IPO Anthropic is still private, but not for much longer if current reporting holds up. The company submitted a confidential IPO filing to the Securities and Exchange Commission in June and has started meeting with potential investors, CNBC reported. Investors expect Anthropic to go public in October at a valuation of $2 trillion or more, which would make it the largest IPO in history, Quartz reported. That is where the Nscale agreement comes in. When Anthropic pitches Wall Street, one of the first questions it will face is whether it can keep growing without running out of the hardware and electricity its Claude models need. This deal is designed to answer that question before anyone asks it. How Anthropic has assembled its computing infrastructure will feature prominently in what it tells prospective public-market investors. What 460 megawatts and Vera Rubin chips actually buy The hardware here is worth understanding, because it explains why the price tag is so large. The West Virginia facility will run on Nvidia's next-generation Vera Rubin chips, which are expected to start coming online at the end of 2027, TechCrunch reported. Vera Rubin is Nvidia's newest system, and it combines six different chips working together at the cutting edge of chip design. The 460 megawatts of power involved is enough to supply about 345,000 US homes at once, according to Bloomberg, which shows the scale of what modern AI training now requires. Here is what the deal locks in for Anthropic: Key terms of the Anthropic and Nscale agreement * Total value: Roughly $45 billion over six years, averaging about $7.5 billion a year * Capacity: Approximately 460 megawatts of power * Location: Nscale's Monarch campus in West Virginia * Hardware: Nvidia Vera Rubin systems, online in late 2027 SOPA Images / Getty Images How this fits Anthropic's wider spending run The Nscale deal is not a one-time move. It follows a series of large agreements Anthropic has signed to lock down capacity. Earlier this month, the company signed a $10 billion, six-year deal with Volta for a data center in Norway, TechCrunch reported. In July, it added a $5 billion computing-related agreement with AMD. And back in May, Anthropic entered a large arrangement with SpaceX that reportedly provides about $1.25 billion worth of capacity each month. There is a reason for all of this activity. Anthropic acknowledged earlier this year that heavy demand for Claude had strained its existing systems and caused reliability problems during peak hours. By spreading its work across Nvidia chips, AWS Trainium, and Google's processors, Anthropic avoids depending on any single vendor. What the numbers say about Anthropic's growth Anthropic surpassed OpenAI in quarterly revenue for the first time in the second quarter, posting $11.6 billion, more than double its first-quarter total, Bloomberg reported. Its annualized revenue run rate climbed from about $9 billion in late 2025 to more than $47 billion in May and near $65 billion by July. The pitch to investors leans on an even bigger figure. Anthropic is preparing to tell IPO investors that its total addressable market tops $30 trillion, based on the full scope of work AI models could perform across industries, the The Wall Street Journal reported. That claim has drawn skepticism. NYU finance professor Aswath Damodaran is widely known as the dean of valuation for decades of work breaking down company worth. He said a similar AI market estimate from SpaceX was "reaching the end of what's plausible and pushing beyond," BetaNews reported. The risks investors should weigh before the IPO First, there is timing. The West Virginia facility will not come online until late 2027, so this is a long-term backup plan rather than a fix for today's capacity shortage. Before then, Anthropic has to bridge the gap using its existing arrangements with Amazon, Google, and SpaceX. Second, there is the question of profit. A $45 billion commitment to one partner adds to already enormous obligations. Investors will want to see that revenue growth can outpace these costs over time, not just for a quarter or two. Third, the valuation itself is unproven. Anthropic's last private raise valued it at $965 billion in May, according to according to Briefs.co, so a $2 trillion debut would roughly double that mark within months. What this means for investors You cannot buy Anthropic shares yet, but you can prepare. If the IPO arrives this autumn, treat the pitch figures with care. A $30 trillion market claim is a projection, not a guarantee, and Anthropic's own forecast of $190 billion to $200 billion in 2028 revenue would capture under 1% of that market. Watch three things once the filing lands: * Whether revenue keeps doubling * Whether adjusted profit holds up under audited numbers * Whether these capacity deals turn into reliable service instead of more strain. The Nscale agreement shows Anthropic can secure the power and chips it needs. Whether it can turn that into durable profit is the question that will decide if a $2 trillion price tag holds up once investors get their first real look. Investors don't need to wait for the IPO to get exposure to this story. Every one of these deals runs on Nvidia hardware, and rivals like AMD are fighting for a share, which keeps AI infrastructure names firmly in focus heading into the debut. The Arena Media Brands, LLC THESTREET is a registered trademark of TheStreet, Inc. This story was originally published August 28, 2026 at 8:33 PM.
Anthropic's $45B cloud deal points to its IPO
Anthropic just made one of the largest infrastructure commitments in AI history, and the timing is hard to ignore. The company agreed to spend roughly $45 billion renting cloud computing power from Nscale, a UK-based AI infrastructure firm. The deal runs six years and covers about 460 megawatts of power at Nscale's flagship data center development in West Virginia. For investors, this matters because Anthropic is preparing to go public, and this agreement tells you a lot about the case it plans to make. Why Anthropic's Nscale deal points straight at its IPO Anthropic is still private, but not for much longer if current reporting holds up. The company submitted a confidential IPO filing to the Securities and Exchange Commission in June and has started meeting with potential investors, CNBC reported. Investors expect Anthropic to go public in October at a valuation of $2 trillion or more, which would make it the largest IPO in history, Quartz reported. That is where the Nscale agreement comes in. When Anthropic pitches Wall Street, one of the first questions it will face is whether it can keep growing without running out of the hardware and electricity its Claude models need. This deal is designed to answer that question before anyone asks it. How Anthropic has assembled its computing infrastructure will feature prominently in what it tells prospective public-market investors. What 460 megawatts and Vera Rubin chips actually buy The hardware here is worth understanding, because it explains why the price tag is so large. The West Virginia facility will run on Nvidia's next-generation Vera Rubin chips, which are expected to start coming online at the end of 2027, TechCrunch reported. Vera Rubin is Nvidia's newest system, and it combines six different chips working together at the cutting edge of chip design. The 460 megawatts of power involved is enough to supply about 345,000 US homes at once, according to Bloomberg, which shows the scale of what modern AI training now requires. Here is what the deal locks in for Anthropic: Key terms of the Anthropic and Nscale agreement * Total value: Roughly $45 billion over six years, averaging about $7.5 billion a year * Capacity: Approximately 460 megawatts of power * Location: Nscale's Monarch campus in West Virginia * Hardware: Nvidia Vera Rubin systems, online in late 2027 SOPA Images / Getty Images How this fits Anthropic's wider spending run The Nscale deal is not a one-time move. It follows a series of large agreements Anthropic has signed to lock down capacity. Earlier this month, the company signed a $10 billion, six-year deal with Volta for a data center in Norway, TechCrunch reported. In July, it added a $5 billion computing-related agreement with AMD. And back in May, Anthropic entered a large arrangement with SpaceX that reportedly provides about $1.25 billion worth of capacity each month. There is a reason for all of this activity. Anthropic acknowledged earlier this year that heavy demand for Claude had strained its existing systems and caused reliability problems during peak hours. By spreading its work across Nvidia chips, AWS Trainium, and Google's processors, Anthropic avoids depending on any single vendor. What the numbers say about Anthropic's growth Anthropic surpassed OpenAI in quarterly revenue for the first time in the second quarter, posting $11.6 billion, more than double its first-quarter total, Bloomberg reported. Its annualized revenue run rate climbed from about $9 billion in late 2025 to more than $47 billion in May and near $65 billion by July. The pitch to investors leans on an even bigger figure. Anthropic is preparing to tell IPO investors that its total addressable market tops $30 trillion, based on the full scope of work AI models could perform across industries, the The Wall Street Journal reported. That claim has drawn skepticism. NYU finance professor Aswath Damodaran is widely known as the dean of valuation for decades of work breaking down company worth. He said a similar AI market estimate from SpaceX was "reaching the end of what's plausible and pushing beyond," BetaNews reported. The risks investors should weigh before the IPO First, there is timing. The West Virginia facility will not come online until late 2027, so this is a long-term backup plan rather than a fix for today's capacity shortage. Before then, Anthropic has to bridge the gap using its existing arrangements with Amazon, Google, and SpaceX. Second, there is the question of profit. A $45 billion commitment to one partner adds to already enormous obligations. Investors will want to see that revenue growth can outpace these costs over time, not just for a quarter or two. Third, the valuation itself is unproven. Anthropic's last private raise valued it at $965 billion in May, according to according to Briefs.co, so a $2 trillion debut would roughly double that mark within months. What this means for investors You cannot buy Anthropic shares yet, but you can prepare. If the IPO arrives this autumn, treat the pitch figures with care. A $30 trillion market claim is a projection, not a guarantee, and Anthropic's own forecast of $190 billion to $200 billion in 2028 revenue would capture under 1% of that market. Watch three things once the filing lands: * Whether revenue keeps doubling * Whether adjusted profit holds up under audited numbers * Whether these capacity deals turn into reliable service instead of more strain. The Nscale agreement shows Anthropic can secure the power and chips it needs. Whether it can turn that into durable profit is the question that will decide if a $2 trillion price tag holds up once investors get their first real look. Investors don't need to wait for the IPO to get exposure to this story. Every one of these deals runs on Nvidia hardware, and rivals like AMD are fighting for a share, which keeps AI infrastructure names firmly in focus heading into the debut. The Arena Media Brands, LLC THESTREET is a registered trademark of TheStreet, Inc. This story was originally published August 28, 2026 at 5:33 PM.