Wall Street is abuzz with anticipation and caution as Anthropic, one of the world’s leading artificial intelligence firms, prepares for its much-anticipated initial public offering (IPO). The company’s meteoric revenue growth—expected to top $100 billion in annual sales by the end of 2026, according to The New York Times—has investors speculating that Anthropic’s enterprise value could soar as high as $2 trillion. But behind the excitement, a complicated web of costs, global trade shifts, and investment concentration is raising as many eyebrows as hopes.
Anthropic’s journey over the past year has been nothing short of extraordinary. Just two months ago, the company’s annual revenue forecast stood at $65 billion. By September 2026, that estimate had rocketed upward by $35 billion, a testament to the breakneck pace of adoption for Anthropic’s flagship AI products, including its widely used 'Claude' suite. The number of enterprise customers spending more than $100,000 annually on Anthropic’s services quadrupled from roughly 1,500 at the end of 2025 to about 6,000 by June 2026. Even more impressive, over 100 major clients are now shelling out more than $10 million each year for the company’s technology.
Big names like AIG, Thomson Reuters, and Spotify have all integrated Anthropic’s AI tools into their operations, leading to measurable gains in productivity and efficiency. These success stories have become central to the company’s pitch to investors as it moves closer to a public listing. Yet, despite these headline-grabbing numbers, there’s a growing sense of caution among Wall Street analysts. As The Information reported, Anthropic’s operating costs remain daunting: in spring 2025, the company was spending $2.30 for every $1 in revenue. It wasn’t until the second quarter of 2026 that Anthropic surprised the market by posting a positive adjusted operating margin, flipping the script on the common perception that AI companies are endless money pits.
But will that profitability last? Anthropic’s management has been upfront—if unofficial—about its plans to aggressively spend its $12 billion war chest on next-generation AI model training and massive data center contracts. The company’s May 2026 deal with SpaceX, worth a staggering $1.25 billion per month for computing infrastructure, ate up two-thirds of its second-quarter infrastructure budget. As these investments ramp up, analysts warn that operating margins could quickly slip back into the red.
It’s not just Anthropic feeling the heat. The entire global economy is being reshaped by the AI boom. The World Trade Organization (WTO) reported that, in the first quarter of 2026, AI-related goods made up about 19% of global merchandise trade, with trade in AI technology products soaring by 42% year-on-year. Electronic components and AI equipment have become the engines of global trade growth, even as other sectors like chemicals, steel, and fuel have seen declines.
Asia is at the epicenter of this transformation. The region’s exports of AI goods jumped 12.9% in early 2026, and imports climbed 14.6%, thanks to bustling intra-regional trade among China, South Korea, Taiwan, Singapore, and Thailand. Meanwhile, North America—especially the United States—dominates the capital side, with 76% of global AI venture investment flowing into American startups and scale-ups. As Financial Times highlighted, this division of labor has only deepened: U.S. firms design and fund the technology, while Asian manufacturers supply the semiconductors and components that power the AI revolution.
The surge in AI demand has triggered a corresponding explosion in capital expenditures. According to LSEG, the combined capex of U.S. tech giants like Microsoft, Alphabet, Amazon, Meta, and Oracle jumped from $485 billion in January 2026 to $730 billion by July. Amazon alone raised its 2026 capital spending plan to $220 billion, with its AWS cloud division reporting a 37% year-on-year revenue increase in the second quarter and a backlog of orders swelling from $364 billion to $496 billion. CEO Andy Jassy put it bluntly: “AWS is growing explosively. Even $220 billion isn’t enough to meet this year’s demand.”
This investment frenzy has its consequences. Amazon’s free cash flow swung from an $18.2 billion surplus a year ago to a $7.6 billion deficit, while Alphabet posted a $5.9 billion negative free cash flow in Q2 2026. Bridgewater estimates that AI infrastructure spending by the top four U.S. tech firms will hit $650 billion this year, up 59% from 2025. “The AI investment race is entering a more dangerous phase,” warned Greg Jensen, Bridgewater’s co-chief investment officer. If projected returns fall short, both investment flows and financial markets could face turbulence.
Energy and minerals are now firmly in the AI supply chain’s crosshairs. The International Energy Agency (IEA) forecasts that global data center electricity consumption will nearly double from 485 TWh in 2025 to 950 TWh by 2030, with AI-specific centers tripling their usage. The power demands are staggering: a single server rack in a top-tier data center could soon consume as much electricity as 65 households. Copper and electrical equipment trade are surging as companies invest in new servers, power grids, and cooling systems. But the IEA’s Fatih Birol cautions that power grid shortages and high electricity prices could slow the pace of new data center construction.
On the investment side, the concentration of capital in AI is becoming more pronounced. The Software Policy Research Institute (SPRi) revealed that AI accounted for 75.9% of global venture capital investment in the first half of 2026, a dramatic rise from 23.5% in 2021. The average AI VC deal size has ballooned from $13.5 million to $55.5 million over the same period. The U.S. continues to dominate, attracting 84.1% of global AI VC investment, with late-stage deals (Series C and beyond) making up nearly 73% of all AI VC funding. In the first half of 2026 alone, there were 34 AI investment deals worth over $1 billion each, and just the top three AI companies captured 55.7% of all global AI VC investment.
The focus of these investments is shifting, too. Whereas AI services and applications once took center stage, foundational AI technologies—especially model companies—now account for 61% of large deals. In South Korea, the trend is similar: AI VC investment now makes up 44.3% of all venture capital, with most of the money flowing to late-stage AI semiconductor and model companies.
Yet, as AI’s economic gravity grows, so do the risks. The WTO projects global merchandise trade volume will rise 1.9% in 2026, with AI investments potentially adding half a percentage point to that growth. But if energy prices remain high or if the AI investment boom proves unsustainable, the impact could be just as large in the opposite direction. “If AI investment turns out to be a bubble, it will be hard to sustain the current trade growth,” warned WTO Director-General Ngozi Okonjo-Iweala.
For Anthropic and its soon-to-be shareholders, the stakes couldn’t be higher. CEO Dario Amodei, who has made AI safety a central part of his public messaging, insists that “slowing down does not mean stopping model training or technological progress.” He argues that going public and embracing market transparency will ultimately make the technology—and the company—safer in the long run. But with so much money, energy, and ambition riding on the outcome, Wall Street and the world will be watching closely as the next chapter in the AI story unfolds.