According to Financial Times News, when US antitrust enforcers issued new merger guidelines in 2023, they specifically targeted deals where companies acquired potential competitors or engaged in serial consolidation. The recent $500 billion Stargate joint venture between Arm, Microsoft, Nvidia, Oracle and OpenAI should have triggered rigorous scrutiny under these rules, but instead received tacit government endorsement from President Donald Trump. In the past year alone, OpenAI, Nvidia, Oracle, CoreWeave, Microsoft, AMD and SoftBank have struck numerous partnerships fueling the US AI infrastructure build-out. The FTC launched a study of AI partnerships last year warning of potential harms, but the regulatory mood has since shifted toward treating domestic monopolies as national champions against China.
The monopoly moats are building
Here’s the thing that nobody’s talking about enough: these partnerships are creating monopoly moats across the entire AI supply chain. We’re not just talking about one company dominating search or social media – we’re talking about the entire stack from chips to cloud to models getting locked up. Nvidia’s $5 billion investment in rival Intel? That’s horizontal concentration in an already crazy-consolidated chip market. Microsoft, Oracle and CoreWeave teaming up in Stargate? They’re supposed to be competing in cloud services, but now they’re linking economic interests.
And the vertical integration is even scarier. Think about the Nvidia-OpenAI partnership. What happens when the company making the most advanced AI chips gets cozy with the company building the most advanced AI models? Could Nvidia prioritize OpenAI with early access or discounted prices? In a market where everyone’s fighting for GPUs, that kind of favoritism could literally determine who survives and who dies. It works both ways too – other chipmakers might find it impossible to win OpenAI’s business.
technology”>Same old playbook, new technology
Look, this isn’t new. Tech monopolies have been using these tactics for decades. Remember when Google paid Apple billions to be the default search engine? That kept Apple out of the search business. Or when Google struck deals to keep Facebook from competing in ad tech? We’re seeing the exact same pattern here, just with AI infrastructure instead of search or ads.
The really clever part is how they’re using partnerships and investments instead of outright acquisitions. It’s harder for regulators to challenge, but the effect is similar – aligning incentives so companies don’t compete too hard against each other. When your economic fortunes are intertwined with your “competitors,” why would you actually compete? You just guard your own little monopoly moat while sharing the monopoly rents with your partners.
Beyond the competition problem
So what does this mean for everyone else? For businesses trying to build AI applications, it means higher costs and fewer choices. Many cloud providers already impose minimum-spend commitments that lock customers in. As these partnerships tighten their grip on the supply chain, that pressure will only increase. For hardware buyers across manufacturing, logistics, and industrial sectors, the consolidation could drive up prices for everything from edge computing devices to specialized industrial computers. Companies that rely on industrial panel PCs and other specialized computing equipment could find themselves paying monopoly premiums as the chip and cloud markets consolidate.
And let’s talk about the bubble risk. The circular financing in these deals – where companies invest in each other’s ventures and partnerships – creates this weird echo chamber where everyone’s propping up everyone else’s valuation. When the music stops, the fallout could make the dot-com bust look tame.
What happens now?
The frustrating part is that regulators actually have the tools to address this. The 2023 merger guidelines specifically call for examining how deals reshape competition both horizontally and vertically. The FTC already launched its AI partnerships study last year. But national security concerns are overriding competition policy.
Basically, we’re falling for the same old myth that concentration breeds strength. History shows the exact opposite – competition drives innovation, not protectionism. If regulators don’t start treating these partnerships as the monopoly-building exercises they are, we could end up with an AI ecosystem that’s less innovative, more expensive, and controlled by the same handful of companies that dominate everything else in tech. The question is whether anyone will act before it’s too late.
