The news: The AI arms race has morphed into a closed loop as Nvidia, OpenAI, Apple, Meta, Intel, and Google invest in each other—or rely on each other—to build and scale massive AI platforms. 

This phenomenon is driving capital spending to unsustainable highs while locking out smaller players. Here are some of the larger deals of 2025:

• OpenAI signed a seven-year, $38 billion deal with Amazon Web Services (AWS), its first partnership with the cloud market leader.
• Nvidia is investing up to $100 billion into OpenAI to supply chips for 10 GW of data centers—then booking that as revenues. It also bought a 4% to 5% stake in Intel for $5 billion to secure foundry access and diversify away from TSMC.
• Meta is dangling $100 million signing bonuses to poach founders from AI startups like SSI. It’s also spending $600 billion over three years on AI, despite lacking a breakout product.
• Apple, unable to catch up, is reportedly leaning on Google’s Gemini at a rate of $1 billion a year for AI search and Siri features.

The bigger picture: The same firms investing in chips, search, and models are also becoming the largest customers for datacenters and compute capacity. This consolidation has broad implications.

• On the plus side, these investments lock in compute capacity and GPU orders and stoke investor confidence. 
• However, they could take time to reap profits, lead to overconfidence that rivals can work together, and stoke bubble fears.

The complex, interconnected funding ecosystem is fragile, with many points of failure. Any disruption—such as hardware shortages or soaring data center costs—can ripple across every player involved. We’re already seeing shortages for memory storage modules, leading to increased prices.

The challenge: The biggest US hardware, cloud computing, and AI companies working together solidifies a national effort. It also consolidates the biggest players, and their funding. Locking out smaller players and neutralizing their growth.

• Startups reliant on OpenAI APIs or Nvidia GPUs now risk pricing pressure, product deprecation, or outright competition from their upstream suppliers.
• Meta’s pursuit of startup founders like SSI’s Daniel Gross—and its $14 billion investment in Scale AI—indicate talent and IP consolidation that could stoke antitrust scrutiny.
• Apple and Google blending together—rather than competing—could stifle the diversity of models and approaches the AI ecosystem needs to stay vibrant.

For competitors, there are fewer paths to scale unless they align with a tech giant—or get absorbed by one. That raises concerns over competition, innovation stagnation, and limited user choice in AI platforms and assistants.

The AI money loop fuels infrastructure buildouts, model training, and short-term investor enthusiasm, but it also risks narrowing the field and inflating expectations. The deeper these companies invest in each other, the harder it may be to untangle if promised breakthroughs don’t materialize.

Looking ahead: For now, this cycle looks virtuous—cash becomes compute, compute becomes product, product becomes revenue. But if users balk, costs rise, or regulators intervene, the loop could snap and expose how a few companies held the keys to the AI market. 

For brands, there’s a lot to unpack. Evaluate your AI stack to understand who really owns the technology and which providers can ensure continuity. Consider alternatives and constantly assess costs as competition grows.

This content is part of EMARKETER’s subscription Briefings, where we pair daily updates with data and analysis from forecasts and research reports. Our Briefings prepare you to start your day informed, to provide critical insights in an important meeting, and to understand the context of what’s happening in your industry. Non-clients can click here to get a demo of our full platform and coverage.