Advanced Micro Devices will sell Anthropic tens of billions of dollars' worth of AI servers and invest as much as $5 billion in the Claude maker—a deal that reveals the staggering financial commitments now required to compete in artificial intelligence. Anthropic will buy up to two gigawatts of AMD's latest-generation Instinct MI450 chips starting in the first half of 2027, locking the startup into a capital-intensive model that mirrors the broader crisis facing Silicon Valley's largest companies.
The spending trajectory is becoming unsustainable. Microsoft, Alphabet, Amazon, Meta Platforms and Oracle—the so-called hyperscalers—are expected to spend more combined on capital expenditures than they generate in free cash flow by 2027, according to a Reuters analysis of LSEG consensus estimates. These five companies will generate about $340 billion more in annual operating cash flow in 2027 than in 2025, but capex is expected to rise by roughly $534 billion. That's $1.57 of additional investment for every $1 of additional cash flow.
Consider what this means for workers, communities, and the broader economy. Massive capital requirements concentrate wealth and power in a handful of corporations. It also raises questions about whether this spending benefits the public or simply enriches shareholders and executives. Current-year consensus estimates for the five companies' capex have risen from about $485 billion in January to around $730 billion in July—a 50% increase in just six months.
The Compute Bottleneck and Market Consolidation
Anthropics's deal with AMD underscores a critical problem: access to computing power is now the limiting factor in AI development, and only the richest companies can afford it. Tom Brown, co-founder and compute head at Anthropic, acknowledged this bluntly: "Access to compute is central to keeping Claude at the frontier and meeting demand from our customers." Anthropic has moved aggressively in recent months to overcome capacity constraints, renting the full computing power of SpaceX's Colossus 1 facility in Memphis in May, which houses more than 220,000 Nvidia processors and gave it 300 megawatts of new capacity. Meta Platforms is also in talks to lease computing power to Anthropic in a potential deal worth up to $10 billion over two years.
These "circular deals"—where chipmakers invest in AI firms that are among their biggest customers—concentrate market power among a shrinking number of players. AMD's investment in Anthropic is tied to achieving certain deployment milestones, meaning the chipmaker has direct financial incentive in the startup's success. Nvidia has been in talks to invest $30 billion in ChatGPT creator OpenAI. In October, AMD announced a multi-year deal with OpenAI that would bring in tens of billions of dollars in annual revenue while giving the ChatGPT creator the option to buy up to roughly 10% of the chipmaker. These arrangements blur the line between competition and collusion, concentrating control over AI development in the hands of a few corporations.
The Free Cash Flow Crisis
The human cost of this spending spree isn't immediately visible on balance sheets, but it's real. Investors are noticing that rapid growth in cloud and AI revenue isn't keeping pace with the spending surge. Hyperscalers led the market rally since the AI buildout began, but over the last year, all but one—Alphabet—have trailed the S&P 500. Oracle's situation is most alarming: its shares have lost 36% so far this year as its free cash flow has turned negative. Oracle's capex as a percentage of operating cash flow rose from 47% in fiscal 2022 to 174% for fiscal 2026, which ended in May. Capex came to $55.7 billion in its most recent fiscal year, compared with operating cash flow of $32 billion. The company plans to raise $45 billion to $50 billion through debt and equity to fund cloud infrastructure expansion.
Microsoft reported $35.8 billion in operating cash flow in its fiscal second quarter while recording $37.5 billion of capital expenditures, including finance leases. Amazon said trailing 12-month operating cash flow rose 30% to $148.5 billion in the first quarter, but free cash flow fell to $1.2 billion. This matters because it means less money for workers' wages, community investment, and public goods.
What Investors Are Demanding
Shay Boloor, chief market strategist at Futurum Equities, said, "Investors are underestimating how fundamentally AI is changing the Big Tech business model." He explained: "These companies were historically valued as asset-light platforms because revenue could scale much faster than capital requirements, but AI is pushing them toward a hybrid model where software, advertising and cloud economics increasingly depend on enormous physical infrastructure spending."
There are some signs that AI spending is paying off. Microsoft has said its AI business has surpassed a $37 billion annual revenue run rate, while Amazon reported 28% growth at its AWS unit in its first quarter. Yet the capex estimates cover all spending, not just AI-related expenditures, because companies do not consistently disclose AI-specific investment. This lack of transparency makes it impossible for workers, regulators, or the public to understand where their economy is headed.
David Russell, global head of market strategy at TradeStation, articulated the core tension: "Earnings growth may not be enough to justify investment if capex is depleting cash. Companies exist to make money, not spend money." Freddy Lavric, senior trader at Winthrop Capital Management, warned: "Over the next two to three years, companies need to show that AI is driving incremental revenue, expanding margins and improving cash flow. If those financial benefits aren't becoming evident by then, the market will start questioning whether the investment cycle has gone too far."
The other companies so far have been able to return cash to shareholders. Microsoft, Alphabet and Meta still generated enough free cash flow to cover dividends and buybacks in their latest fiscal years, according to SEC filings. But buybacks could be at risk if spending remains elevated and AI monetization takes longer than expected. When those companies report earnings, investors will be watching closely for evidence that the AI buildout is actually generating returns—or whether it's simply consuming capital without delivering broad-based prosperity.
Why This Matters:
The AI spending surge reveals a fundamental imbalance in how our economy allocates resources. Five corporations are spending hundreds of billions of dollars annually on infrastructure that will determine the future of work, information, and economic opportunity—yet these decisions remain entirely private. Workers have no voice in whether their jobs are being automated. Communities have no say in where data centers are built or what environmental costs they bear. The public has no transparency about AI development timelines or capabilities. Meanwhile, free cash flow constraints mean less money for wage increases, benefits, and public investment. If the current trajectory continues and AI monetization doesn't materialize, we'll have spent trillions on infrastructure controlled by a handful of companies, with little to show for it in terms of shared prosperity or democratic accountability. The spending estimates themselves have doubled in just six months—suggesting that neither investors nor executives truly understand what they're building or whether it will work.