Amazon’s AI and satellite spending spree sparks investor jitters amid cloud arms race

Amazon's $11.6 billion acquisition of Globalstar marks a pivotal expansion into satellite direct-to-device (D2D) services, strategically complementing its broader $200 billion AI infrastructure investment. This move intensifies the competitive landscape against SpaceX's Starlink by reshaping the satellite communications and cloud compute markets, signaling bold shifts in capital allocation and digital transformation focused on next-generation connectivity and logistics automation.

Amazon's aggressive investments in AI startups, particularly its $25 billion commitment to Anthropic, underscore a high-stakes strategy to dominate the AI cloud ecosystem. Securing a $100 billion AWS cloud spend and integrating Amazon's Trainium chips, this partnership echoes Amazon's earlier collaboration with OpenAI and reflects a seismic shift in AI cloud infrastructure, compute capacity, and strategic capital deployment amid the intensifying AI arms race of 2026.

AWS is experiencing robust AI-driven growth, with revenue surging 28% year-over-year to $37.6 billion and operating margins near 38%, fueled by a rapidly expanding custom chip business that has surpassed a $20 billion annualized revenue run rate. By undercutting NVIDIA’s AI training costs, Amazon is redefining cloud and AI economics, shifting from GPU-centric training to agentic AI-powered cloud services, and positioning itself as a dominant AI infrastructure powerhouse leveraging its vertically integrated chip and cloud strategy.

Amazon's AI silicon business has rocketed to a $50 billion run rate, fueling a cloud AI arms race where innovations like stateful agent platforms and supply-driven cloud migrations are transforming enterprise AI workflows and operational efficiency. This growth, combined with a 67% rise in capital expenditures to $147.3 billion in 2026, highlights Amazon’s commitment to locking in massive AI spending and infrastructure control, despite investor concerns over debt and margin pressures.

Amazon's aggressive $200 billion AI infrastructure investment, including its $13 billion commitment to Anthropic tied to over $100 billion in AWS service spending, exemplifies a circular funding model that blurs traditional lines between capital infusion and service procurement. While this strategy aims to secure long-term compute resources amid soaring AI demand, it has fueled investor skepticism due to the opaque nature of these deals and uncertainty about the actual cash versus in-kind contributions involved, as highlighted by concerns over transparency and deal sustainability.

Financially, Amazon continues to demonstrate solid growth and profitability, with AWS revenue surging 28% to $37.6 billion and a backlog exceeding $460 billion, driven by AI partnerships with OpenAI and Anthropic. However, this growth is overshadowed by rising debt levels and a dramatic plunge in free cash flow—from $26 billion to just $1.2 billion over the trailing 12 months—due to a $59.3 billion year-over-year increase in capital expenditures for data centers and satellite infrastructure, raising concerns about margin pressures especially in retail and advertising segments.

Investor sentiment remains cautiously optimistic yet tinged with skepticism, as market participants debate whether Amazon’s massive AI and cloud investments will translate into sustainable high-margin growth or devolve into a capital-intensive, lower-margin business. Analysts like Dan Caplinger liken Amazon’s current out-of-favor tech stock status to Alphabet’s position years ago, emphasizing uncertainty about the timing and certainty of returns despite AWS’s strong AI-driven growth and disciplined margin improvements under CEO Andy Jassy.

Amazon’s strategic choice to forgo stock buybacks and dividends preserves financial flexibility, allowing it to fund its expansive AI infrastructure buildout without exacerbating leverage, even as free cash flow remains compressed near zero. This deliberate investment in building a comprehensive enterprise AI stack—from custom chips to managed AI agents—reflects a long-term vision that, while unsettling some investors in the near term, positions Amazon to capitalize on the burgeoning AI cloud market and avoid the bigger mistake of missing this critical growth inflection.

Amazon’s bold $11.57 billion acquisition of Globalstar marks a pivotal escalation in the orbital broadband and AI infrastructure arms race, directly challenging SpaceX’s Starlink dominance. By integrating satellite communications into its AI cloud ecosystem, Amazon is not only expanding its next-generation connectivity capabilities but also signaling a strategic shift in capital allocation and digital transformation priorities that could redefine the competitive dynamics of cloud compute and satellite networks.

The rivalry between AWS and Google epitomizes the broader AI cloud infrastructure battle heating up amid a projected trillion-dollar AI boom through 2027. Both giants are aggressively leveraging custom-designed chips and forging strategic partnerships to capture market share, underscoring how hyperscalers are racing to build the most powerful, efficient AI compute platforms. This fierce competition reflects the high stakes and rapid innovation fueling the AI cloud arms race, with Microsoft and others also jockeying for position in this lucrative and transformative market.