TDC 069: The Circular AI Money Shell Game?? | What To Do About The AI Bubble That May Soon Be Coming To an End

TDC 069: The Circular AI Money Shell Game?? | What To Do About The AI Bubble That May Soon Be Coming To an End

The AI bubble may be closer to popping than you think—and the evidence is hiding in plain sight.

Episode Summary

In this episode of The Digital Contrarian, host Ryan Levesque goes into the current state of the AI bubble and where we might be on the hype cycle curve.

You'll learn why enterprise AI adoption is actually declining, discover the circular money shell game happening between major AI players, and uncover a potential opportunity hiding in OpenAI's recent job postings that could rival the early days of Facebook Ads.

Question of the Day 🗣️

Where do YOU think we are on the AI bubble curve? Share your perspective in the comments below!

Key Take-aways

Enterprise AI adoption has fallen from 14% to 12% between June and August 2025—unprecedented for any new technology
The NASDAQ curves from the dot-com bubble (1998-2001) mirror today's AI-driven market almost identically
Circular AI deals between OpenAI, Nvidia, and AMD echo the financial engineering that preceded past bubble crashes
A single ChatGPT algorithm change wiped out $8 billion in Reddit's market value by reducing citations 86%
OpenAI's recent job postings suggest an ad-supported model may be coming sooner than expected

Timestamped Outline ⏱️

00:00 – The AI bubble question everyone's asking
00:08 – Why you can't see bubbles from the inside
01:40 – Where are we on the AI hype cycle curve?
02:12 – The late-stage bubble evidence piling up
03:04 – The circular AI money shell game explained
04:07 – Why 95% of enterprise AI pilots are failing
04:42 – The anti-bubble argument (and why it matters)
06:41 – Unprecedented volatility: AMD up 35%, Reddit down 22%
07:36 – The neuroscience of staying calm in chaos
08:23 – OpenAI's next move: The biggest opportunity since Facebook Ads?
08:59 – New agency service announcement

Links & Resources 🔗

Issue #068 of The Digital Contrarian – "Why Your $20/month AI tool is about to cost $500..." → https://ryanlevesque.net/your-20-month-ai-tool-is-about-to-cost-500/

Issue #039 of The Digital Contrarian – "Right Thesis, Wrong Timing?" → https://ryanlevesque.net/right-thesis-wrong-timing/

Derek Thompson's "This Is How the AI Bubble Will Pop" → https://www.derekthompson.org/p/this-is-how-the-ai-bubble-will-pop

Bloomberg's circular AI money flow diagram → https://www.bloomberg.com/news/articles/2025-10-08/the-circular-openai-nvidia-and-amd-deals-raising-fears-of-a-new-tech-bubble

Return to Real Agency waitlist → https://go2.bucketforms.com/sf/8a8a42b9

1:1 VIP Client waitlist → https://ryanlevesque.net/1-1-vip-waitlist/

Connect & CTA 🎯

👉 Enjoyed this? Subscribe & leave a review on Apple Podcasts.
🎁 Join 100 000+ digital entrepreneurs who get Ryan's "Strategic Insights for Digital Entrepreneurs Who Think Differently" every weekend: https://ryanlevesque.net/join-the-digital-contrarian/

Credits

Host: Ryan Levesque
 © 2025 RL & Associates LLC. All rights reserved.

Transcript

The circular AI money shell game and what to do about the AI bubble that may soon be coming to an end. Now the annoying thing about bubbles is that in the moment there is no way of knowing for sure that you are actually in a bubble or more accurately knowing which phase of the bubble you're in. A classic case of being inside the bottle and not being able to read the label. 

Now last week I wrote a piece titled issue number 68 why your $20 a month AI tool is about to cost $500 and it created quite an uproar. Many expressed appreciation for the heads-up, some dismissed the possibility entirely, and one reader was outraged, accusing the piece as being clickbait. All of these reactions, by the way, represent the expected range of reactions from delusion to denial while in the middle of a bubble, for what it's worth.

But here's the thing, whether you agree or disagree with the possibility of a major price hike in generative AI to support the cost structure of the technology, one of my objectives with the Digital Contrarian Newsletter is to offer an informed contrarian perspective that makes you think, to help you spot non-obvious trends and patterns and forces that others may be completely missing, to see hidden opportunities, to be aware of risks lurking in the shadows, and most importantly, to connect dots in a way that helps you make better decisions that impact the future of your business and in turn your life. Something private clients, by the way, pay me to serve in that role as their strategic advisor. And to that end, I believe a question we should all be thinking about right now is this, where on the AI bubble curve in the image are we right now at this moment? And more importantly, how should we be preparing for what's about to come next? Well, first things first, let's take a look at some of the latest evidence suggesting that we may be closer to the top of this bubble than some may think, which brings us to the late stage bubble argument.

Now, on the one hand, one could argue that the evidence suggesting that we're approaching the top of the AI bubble is already quite significant and growing. From the now near identical curves of the NASDAQ during the dot-com bubble of 1998 to 2001, versus the NASDAQ today, 2023 to 2025, driven by AI hype. First shared by Spencer Hakimian of Tulu Capital.

To the circular AI money shell game taking place over the last few weeks, like in the diagram first reported by Bloomberg here, and how similar forms of extreme financial engineering have been a last gasp move at the center of each of the most recent financial bubbles. Think collateralized debt obligations in 2006 to 2007, leading to the 2008 housing market crash and ultimately the subsequent global financial crisis. And this is evidence that is just the tip of the iceberg. 

For example, for the first time ever, the US Census Bureau data has reported that 1.2 million firms that have been surveyed have shown that enterprise AI adoption has fallen from 14% to 12% between the months of June and August of 2025. As Stephen Klein recently wrote, this has never happened with any technology in early adoption. Not electricity, not the Internet, not mobile phones.

And just this past week, we've seen evidence that the usage of AI vibe coding tools are now also in the decline. As Barclays analysts have reported, Lovable hit $100 million in annual recurring revenue back in June of this year, but since then traffic has dropped by 40%. For cells, VO got hit even harder. 

Visits are down 64% since May. Bolt.new is down 27% since June and even Repl.it, one of the stronger players, has seen traffic slip. And by the way, Google Trends data backs all of this up.

And by the way, this is all in addition to some of the data that we've explored in past issues of The Digital Contrarian over the last several weeks, from issue number 62, the MIT McKinsey study, that shows that 95% of generative AI pilots are failing to deliver measurable ROI at the enterprise level. From issue number 68, OpenAI is predicted to lose $83 million per day between now and 2029, with no clear profitable revenue model in sight. So again, one could argue that the evidence that we may be approaching the top of the AI bubble is already quite significant.

But that brings us to the anti-bubble argument. But on the other hand, an argument can be made that the louder the bubble talk gets, the less and less likely that we've actually reached the delusion peak stage before the crash. After all, as we explored in issue number 39, Right Thesis, Wrong Timing, Alan Greenspan made his famous irrational exuberant speech predicting the dot-com bubble in 1995, nearly five years before the eventual stock market crash.

So perhaps there's still a long, long way to go. Now, remember, the annoying thing about bubbles is that you can never know for sure when you are inside of one, or know which phase of the bubble that you're in. Hindsight, of course, is 20-20, and there are always telltale signs in retrospect, but in the moment, it's impossible to know for sure.

So if that's the case, the question that we might be asking ourselves is, well, what do we do about all of this based on where we are at this moment? Well, first, I think it's important to note that bubbles aren't inherently bad. In fact, bubbles are a necessary element in the process of bringing forth any form of world-changing technology to humanity. We experienced this with railways in the 1870s, telecom in the 2000s, shale in the 2010s, and now with generative AI in the 2020s.

By the way, if you're interested in going deeper on this topic, Derek Thompson of The Atlantic wrote an excellent piece on his substack titled This is How the AI Bubble Will Pop, based on an interview he conducted with Dr. Paul Kondrowski, a notable venture capitalist and fellow at MIT's Institute for the Digital Economy. We'll make sure to include a link along with this episode. And I think that once we appreciate that bubbles are a necessary part of every new technology cycle, they're somehow less scary.

And second, I think it's also important to acknowledge that this bubble brings with it unprecedented levels of volatility that we should take into account. Like, for example, take two companies, AMD and Reddit. In the last week, AMD's market value increased by 35%, based entirely on the news of a partnership deal with OpenAI.

At the same time, Reddit's market cap dropped nearly 22% over nearly the exact same period. So what's the story with Reddit? Well, check out this chart. ChatGPT made a single change, due in part to a Google algo change, that reduced the percent of Reddit citations showing up on ChatGPT by 86%.

The result? A massive sell-off of Reddit's stock, wiping out over $8 billion in market value practically overnight. So these are massive swings. They're massive swings up, massive swings down.

And when we experience, or even just hear about, massive swings like this, we can feel unstable. We can feel ungrounded, like the world as we know it is literally shifting beneath our feet. And in some ways, it is.

But in moments like this, I remind myself, be curious, ask questions, and seek truth. Because here's the thing, curiosity is calming. In fact, when you're curious, your brain shifts away from negative emotions and threats and anxiety to learning and growth.

In fact, the neuroscience behind this is incredibly compelling. Questions bring clarity. After all, the entire ask method is built upon this premise.

Change your questions, change your life. And last but not least, the faster you get to the truth, the faster you get to the cash. Like, for example, going back to the very beginning of our conversation, maybe OpenAI is going to increase prices by 25x to cover their costs.

Or maybe they're going to release an ad-supported version of ChatGPT. And maybe this is going to be one of the biggest digital marketing opportunities since Meta launched Facebook ads way back in 2007. And by the way, there's some recent clues that suggest this might be coming sooner than we think.

Take a look at some of the most recent OpenAI job postings and tell me what story they tell. Alright, I'll leave you with that for now. Have a great rest of your weekend.

Remember to hug the ones that you love. And until next week, I wish you all my best. Oh, wait, one last thing.

In case you missed it, last week, we launched a brand new agency service in my business that filled up literally the days that we opened up the doors. Now, in this program, as a client, we help you become what's called a category of one business and establish yourself as the clear and obvious choice in your market. Now, we do this by working with you to take a single weekly source of truth piece of content, like this newsletter that I'm reading right now and speaking to you, designed to convey your category of one thinking and positioning.

We then transform that original thought leadership piece of content into an entire strategic content ecosystem so that you show up everywhere in front of your target audience online. All you do is write one email each week and our team does all the rest. We transform that email to 37 different pieces of content across a dozen different channels and handle all the posting in all these different places, everything from YouTube to LinkedIn, all while preserving your authentic voice.

In fact, this is the exact system that I developed to get my own authentic work like this out into the world, my own unique point of view by writing just one email each week. Now, this program oversold the day it went live, but because there's been so much demand ever since we announced it, we decided to create a waitlist for those who may be interested in working with us the next time that we open up Spot. So if you're interested in being notified, as soon as we have capacity to take on more clients, you can get on the priority waitlist by visiting the link that will include with the description of this episode.

Okay. I'm really going to leave you with that for now. Have a great rest of your week.

Remember to hug the ones that you love and until next time, I wish you all my best. Take care and talk soon.