top of page
Search

Is There an AI Bubble?

  • Writer: Cathy Campo
    Cathy Campo
  • Nov 23, 2025
  • 4 min read

By: Sabari Sunil


Everyone at Kellogg is usually talking about AI, hyped about all the possibilities and recent developments. The famous Wall Street Journal Article about how our amazing professors are using AI for casing or how Nvidia added $1 trillion dollars of market value in the past 4 months. It feels like AI is everywhere, and in many ways, the excitement is justified. Most of the industries we’re targeting post-MBA are already being reshaped by AI.



But as business school students, we should also force ourselves to examine the contrarian viewpoints. Scrutiny—not just hype—should be shaping our perspectives about the future.


Circular Nature of Investments


Let's take a look at a tiny snippet of investments in the AI industry.


Pre 2025: Billions, not Hundreds of Billions



2025: The Shift to Hundreds of Billions


Two things feel obvious:


  1. The investment dollar values have increased by an order of magnitude, and it doesn’t look like this trend is going to slow down.


  2. The same dollars keep flowing around the same ecosystem, starting with Nvidia, flowing through OpenAI and Oracle and going back into Nvidia. How does this help? OpenAI is able to get more VC funding, Oracle and Nvidia’s stock prices are going to the moon. Could this be a case of round-tripping? 


For the non-finance people here (like me), I’ll save you an LLM search:


“Round tripping is when companies recycle investments or spending through each other, creating the illusion of revenue growth or demand by moving money in a closed loop rather than generating true external sales or profits.”

I’ve only summarised a few investments here. The full scale of all the investments and their circular nature is headache inducing. Check out this video that does a good job of diving into the details.


Note: This Bloomberg News infographic is now outdated, Nvidia is worth over $5T.
Note: This Bloomberg News infographic is now outdated, Nvidia is worth over $5T.

Hopes, Dreams, and… Demand?


A common counterargument against the bubble is that there is significant demand for Nvidia GPUs which is fuelling the growth. But it's fair to ask: Where is the demand coming from? And what is the end objective of these AI companies? 


Sam Altman, OpenAI CEO, recently said that OpenAI is committed to developing 30 gigawatts of computing resources for $1.4 trillion and eventually, he would like OpenAI to be able to add 1 gigawatt of compute every week. Each gigawatt of compute costs more than $40 billion dollars.


Why are investors and the U.S. Government pumping so much money into AI infrastructure?

Is everyone afraid of missing out on the next big thing? Is AGI (Artificial General Intelligence) the next big thing? Do we have any concrete proof that AGI is around the corner? For all we know, AGI will only be possible in the next 50 years. How does this affect these inflated valuations? Do I have answers to these questions? Nope. Does anyone?


So… Where Are the Profits?


OpenAI is not a public company yet, but analysing other Big Tech firm’s filings can help us get a better idea of what’s going on with the company. I would highly recommend watching this analysis of Microsoft’s recent 10-Q filing with respect to OpenAI. Fair warning: the video will feel like an accounting case. For those dealing with ACCT trauma here’s a TLDW:


  • In Q3 2025 alone, Microsoft lost $4.1B from their investment in OpenAI.

  • Microsoft holds a 32.5% stake in OpenAI.

  • So, OpenAI had losses amounting to approximately $12B in the most recent quarter.

  • OpenAI CFO, Sarah Friar, suggested at a WSJ conference that OpenAI was exploring the possibility of federal loan guarantees to help finance a massive AI infrastructure expansion.

  • Is the writing on the wall clear? Has OpenAI realized that if they want to drive up investment by another order of magnitude, then the American taxpayer is the only source for it?


Nvidia’s 10-Q filing and earnings call for the last quarter is expected around November 20.


Conclusion


After reading this you might think that I’m spewing sacrilege as an MBAi student. I’m just as hyped about AI as everyone else. The main intent of this article is to analyze whether the current push for investments in AI is too soon to give profitable returns. 


Did the additional readings linked in this article feel like prepping for a strategy class? Good, we should all be discussing the current growth in AI a bit more critically. So, the next time you hear news like Google planning AI data centers in space, ask yourself whether you should be challenging the narrative or buying into the hype.


More Sources! Yay?


I started writing this article wanting to cover some of the technological challenges as well but this article is long enough already. But for those curious here’s a good video summary.


 
 
 

Comments

Subscribe

Submit

Interested in writing for the Kelloggian?

Sponsor

Current Sponsors

cupitol-logo.png
Tim-Calkins.avif
bottom of page