Artificial intelligence is driving one of the biggest technology booms since the early 2000s, but beneath the hype lies a complex web of interdependence. The world’s largest AI firms and chipmakers are increasingly funding each other, creating what some analysts describe as a financial echo chamber. It is a structure that fuels rapid growth but also raises red flags reminiscent of the dot-com bubble.
A Loop of Investment and Demand
At the centre of this ecosystem are a few dominant players: OpenAI, Nvidia, AMD, Broadcom and CoreWeave. In recent months, these companies have forged partnerships that blur the lines between customer, supplier and investor. Nvidia announced a plan to invest up to 100 billion US dollars in OpenAI to help build next-generation models, in exchange for OpenAI constructing massive data centres powered by Nvidia chips. The investment effectively guarantees Nvidia a long-term buyer for its hardware.
OpenAI has since partnered with AMD to diversify its chip supply, gaining an option to buy a 10% stake in the company. It has also teamed up with Broadcom to design custom accelerators, embedding itself deeper into the semiconductor supply chain. Meanwhile, a 6.5 billion US dollar deal with CoreWeave gives OpenAI access to cloud infrastructure that runs on Nvidia hardware, hardware from a company that already holds a stake in CoreWeave.
The result is a closed financial loop where capital, infrastructure and demand continually circulate between the same few firms. Nvidia backs CoreWeave, CoreWeave hosts OpenAI, and OpenAI buys Nvidia chips. Each transaction reinforces the others.
Red Flags and Echoes of the Dot-Com Bubble
While this tightly knit ecosystem offers strategic advantages such as predictable demand, faster innovation and reduced reliance on outside capital, it also poses serious risks. When growth is sustained by mutual investment rather than independent market demand, valuations can become inflated.
Economists have compared the pattern to the dot-com boom of the late 1990s, when technology companies invested in one another, propped up by hype rather than profit. When market confidence wavered, the entire network collapsed in on itself. In AI, similar dangers may be forming. If one company falters, whether from regulatory action, supply shortages or a failed product, the shock could ripple across the entire sector.
This concentration of financial and technical power also limits competition. Smaller firms, researchers and startups may find themselves locked out of the infrastructure needed to train large-scale AI models. With compute and chip access concentrated among a handful of players, the promise of open innovation could give way to a closed club.
Environmental and Ethical Costs
The circular funding model is also fuelling a rapid expansion of energy-intensive data centres. These facilities already consume more than 4% of US electricity and vast quantities of water for cooling. Plans to build new coal-powered plants to meet AI’s energy demand have prompted environmental concerns and accusations of greenwashing.
Critics argue that while AI companies champion efficiency and innovation, their growth model increasingly mirrors the resource-heavy industries they once promised to disrupt. Instead of promoting sustainability, the sector risks reinforcing extractive and wasteful patterns.
Lessons for the Future
History suggests that unchecked optimism rarely ends well. Like the dot-com era, today’s AI boom is fuelled by extraordinary technological promise and equally extraordinary financial entanglement. True progress, analysts warn, depends not on recycled capital but on genuine value creation, innovation that stands on its own rather than being propped up by its partners’ balance sheets.
As the AI economy continues to expand, the question remains whether this circular system represents resilience or risk. The answer may determine not just the future of artificial intelligence but the stability of the wider tech industry itself.
