AI supply constraints prevent near-term bubble
The guest argued that the AI sector is currently supply-constrained rather than demand-constrained, making a near-term bubble unlikely.
The argument
The discussion highlighted severe shortages in compute, memory, power, and data center capacity, with capacity locked up until late 2028 or early 2029. The guest suggested that this supply bottleneck acts as a healthy check on speculative excess, though algorithmic breakthroughs leading to smaller, less token-consumptive models could eventually alter this dynamic.
The thesis, stress-tested
✓ What validates it
- ✓Top model companies reaching a combined $200 billion revenue run rate by year-end
- ✓Data center capacity remaining fully booked through 2028
▸ Risks discussed
- ▸Algorithmic breakthroughs leading to much smaller, less token-consumptive models that reduce hardware demand
- ▸Persistent local and regulatory resistance to data center construction
Hear it yourself
"The way that we thought about much of the AI work that was happening before that was a sort of, like, nebulous promise in the enterprise, but we probably were contextualizing it around things like the cloud and software companies and productivity enhancement. And then on the consumer side, you could think about AI companies like a consumer business, how many users they have Yeah. And what the price is and how big that can get. And by the way, I think that's gonna be much bigger than people expect too, which we could talk about."
AFFILIATE LINK · ZORTIX MAY EARN A COMMISSION · NEVER A RECOMMENDATION TO TRADE